Ann Thorac Surg 1996;61:184-189
© 1996 The Society of Thoracic Surgeons
Original Articles: General Thoracic
Double-Lung Transplantation in the Rat: An Acute, Syngeneic In Situ Model
Bernard Hausen, MD,
Stefanos Demertzis, MD,
Frank Schröder,
Maike Beuke,
Hans-Joachim Schäfers, MD
Division of Thoracic and Cardiovascular Surgery, Surgical Center, Hannover Medical School, Hannover, Germany
Accepted for publication August 24, 1995.
 |
Abstract
|
|---|
Background. The high rate of reperfusion injury in clinical lung transplantation mandates significant improvements in lung preservation. Innovations should be validated using standardized and low-cost experimental models.
Methods. The model introduced here is analyzed by comparing global lung function after varying ischemic times (2, 4, 8, 16, and 24 hours). A rat double-lung block is flush-perfused, and the main pulmonary artery and left atrium are connected to the left pulmonary artery and vein of a syngeneic recipient using a T-shaped stent. With pressure side ports and incorporated flow crystals, measurement of vascular resistance and graft oxygenation can be performed. The transplant is ventilated separately, and compliance and resistance are determined.
Results. The increase in the ischemic interval from 2 to 24 hours caused an increase in the alveolar arterial oxygen difference from 220 ± 20 to 600 ± 34 mm Hg, pulmonary vascular resistance from 198 ± 76 to 638 ± 212 mm Hg•mL-1•min-1, and resistance to airflow from 274 ± 50 to 712 ± 30 cm H2O/L H2O, and a decrease in pulmonary compliance from 0.4 ± 0.05 to 0.12 ± 0.06 mL/cm H2O.
Conclusions. This in situ, syngeneic rat lung transplantation model offers an alternative to large animal models for verification of lung preservation solutions and for modification of donor or recipient treatment regimens.
|
Introduction
|
|---|
Clinical lung transplantation has evolved from an experimental operative procedure to a realistic therapeutic option for patients with end-stage pulmonary vascular and parenchymal disease [1]. Despite numerous experimental innovations regarding various aspects of lung preservation there have been only very limited modifications of the currently used technique in clinical lung transplantation. The incidence of ischemia-reperfusion injury is still alarming and can substantially threaten the success of the transplant procedure [2]. The occurrence of this complication is estimated to be approximately 15% to 40% of all lung transplantation procedures [3].
The majority of investigative approaches dealing with the problem of lung preservation and reperfusion injury [4] have been conducted using either allogenic large animal or ex situ syngeneic rodent models. In theory the basic prerequisites for reperfusion models should be to obtain sufficient and substantial data for online monitoring of graft function, to simulate the actual clinical transplant situation as closely as possible, to be easily reproducible, and preferably to be syngeneic to exclude interference by immunologic events. Most experimental designs introduced so far meet most, but seldom all of these prerequisites. As often encountered in experimental surgery, the simulation of human pathophysiology proves to be extremely difficult. Although large animal models lack syngenicity but offer full coverage monitoring and presumably the closest resemblance to the human physiology, rat models can be syngeneic, easily reproducible, and of low cost [5]. The anatomic dimensions of the rat may, however, limit the in situ concept, the closest resemblance to the ``natural situation.'' The recently developed model described in this article attempts to offer a feasible alternative in experimental surgery related to lung transplantation. Validation of this model is achieved by comparing lung function after increasing ischemic intervals.
|
Material and Methods
|
|---|
Female Lewis rats (350 to 400 g) were obtained from Charles River, Salzfeld, Germany. 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 85-23, revised 1985).
Lung Donor
The Lewis donor rats are anaesthetized with pentobarbital (1 mg/kg body weight intraperitoneally), intubated, and ventilated with a Harvard volume-controlled respirator (Harvard Rodent Ventilator Model 683, South Natick, MA) at 5 mL tidal volume, respiratory rate of 55 breaths per minute, and a positive end-expiratory pressure of 3 cm H2O. A median laparotomy and bilateral thoracotomy are performed. After intravenous application of 100 IU of heparin the apex of left and right ventricle is resected. A large-bore 13-gauge cannula is inserted into the main pulmonary artery, and flush perfusion with reversed Euro-Collins solution is initiated (20 mL in 30 seconds; 4°C temperature; perfusion pressure, 10 cm H2O). The ionic composition of this solution is as follows: Na+, 115 mmol/L; K+ 10 mmol/L; Cl-, 15 mmol/L; PO4-, 57.5 mmol/L; HCO3-, 10 mmol/L; glucose, 3.5 mg/dL; osmolarity, 380 mOsm/L. The perfusion solution is vented through the previously performed left ventriculotomy. At the end of perfusion the lungs are gently wrapped in a soaked gauze and covered with ice slush. Preparation of the graft includes excision of the main pulmonary artery in toto with the subvalvular right ventricular musculature, closure of the mitral valve with 10-0 Prolene (Ethicon, Somerville, NJ) suture, and incision of the left atrial appendage. A T-shaped plastic stent with an internal diameter of 1.8 mm and an overall length of 15 mm with an integrated side port is inserted into the right ventricular outflow tract and ligated. The double-lung block is then excised from the thorax and posterior mediastinum with division of trachea at the cricoid level.
The lungs are ventilated briefly to check for air leaks and then weighed. Finally the lung block is stored at 4° to 6°C in the flush perfusion solution in a fully inflated state (26 cm H2O airway pressure) with the trachea clamped.
Recipient
The Lewis recipient rats are anesthetized with pentobarbital (1 mg/kg body weight intraperitoneally). These animals are then intubated and ventilated, and anesthesia is maintained with isoflurane inhalation (0.3% to 1% volume) at an inspired oxygen fraction of 1.0. For ventilation a pressure-controlled ventilator is used (Infant Ventilator Mark 4; East of Oxford, Oxford, UK) with a gas flow of pure oxygen at 2 L/min, a peak pressure of 20 cm H2O, and a positive end-expiratory pressure of 3 cm H2O. After a left lateral thoracotomy the left lung is mobilized, and the left pulmonary vein and artery are isolated. The bronchus is occluded with a metal clip, cut, and dissected from the surrounding structures of the left lung.
The left pulmonary vein is temporarily clamped as close to the left atrium as possible, and the left pulmonary artery is clamped at the division of the main pulmonary artery. Both vessels are then incised close to the left lung and rinsed to remove the blood. A T-shaped stent with a built-in Doppler probe (H-2R probe; Transonic Systems Inc, Ithaca, NY) (Fig 1
) and a side port for blood gas retrieval and pressure measurement is inserted and ligated. The T stent of the donor double-lung block is now inserted into the recipient pulmonary artery.
View larger version (12K):
[in this window]
[in a new window]
|
Fig 1. . Pulmonary venous T stent for connection of the donor pulmonary vein to the recipient pulmonary vein. Two crystals for Doppler measurement of flow through the stent are integrated. A side port is used for pressure measurements and withdrawal of blood into 100 µL capillaries for blood gas determinations.
|
|
The stent with built-in probes (see Fig 1
) in the donor left pulmonary vein is inserted into the left atrial appendage of the graft and secured. The donor trachea is intubated with a 13-gauge cannula. The donor double-lung block is ventilated with a Harvard volume-controlled respirator (Harvard Rodent Ventilator model 683) at 5 mL tidal volume, a respiratory rate of 50 to 65 breaths per minute, and a positive end-expiratory pressure of 3 cm H2O. The inspiratory oxygen fraction is 1.0. The side ports of the stents are connected to pressure transducers and the flow probe to a flowmeter (Transonic HT 107). The clamps on the recipient pulmonary vein and artery are removed and 100 IU heparin is injected (Fig 2).
View larger version (51K):
[in this window]
[in a new window]
|
Fig 2. . Experimental setup of the acute double-lung transplantation model. A donor lung is connected to a syngeneic recipient animal with two specially designed stents. The donor lung is ventilated separately. A Fleisch tube with analog/digital (A/D) converters allows isolated determination of donor lung function. The A/D converter is connected to a personal computer (PC) for online data recording and display. The side ports of both stents are used for pressure and blood gas measurements. (LAP = left atrial pressure; PAP = pulmonary artery pressure.)
|
|
The respiratory rate of the donor lung is adjusted according to left pulmonary venous blood gas analysis. The target carbon dioxide tension is 30 to 40 mm Hg. Fluid resuscitation is achieved with either blood or crystalloid fluid introduced through the pulmonary artery stent to ensure a pulmonary arterial pressure of 20 mm Hg and a hematocrit of 30% to 40%. The donor lung is kept moist by intermittent topical application of warm fluid. The pH measured in the pulmonary venous blood gases is titrated with 8.5% NaHCO3 to achieve a pH of 7.35 to 7.45.
Measurements
The observation period for these experiments has been limited to 120 minutes, with serial measurements beginning 5 minutes after removal of the pulmonary artery clamp and then repeated every 20 minutes. Each measurement includes the following parameters: left pulmonary arterial and venous blood gas analysis (Corning model 178 blood gas analyzer; Ciba Corning Diagnostics GmbH, Fernwald Germany; measured parameters are oxygen tension, carbon dioxide tension, pH, HCO3, oxygen saturation, and base excess), left pulmonary arterial and venous blood pressure (mm Hg; Hellige Recomed; Hellige GmbH, Freiburg, Germany), isolated donor lung blood flow (mL/min; Transonic flowmeter HT107 connected to flow probe built inside the pulmonary venous tent/H-2R probe, and donor lung function (Pulmonary Monitoring System; Mumed Ltd, London UK), measured by continuous recording of transpulmonary pressure and tracheal flow with pressure transducers and pneumotachograph connected to an analog/digital converter and a personal computer. Measured parameters include resistance (cm H2O•L-1•s-1) and dynamic compliance (mL/cm H2O). Derived parameters are alveolar-arterial oxygen difference [(713 x inspired oxygen fraction) - (carbon dioxide tension - oxygen tension)] and pulmonary vascular resistance [(pulmonary arterial pressure - pulmonary venous pressure)/donor lung blood flow). Two minutes before each measurement a sigh ventilation of the donor lung is performed. Blood for blood gas analysis is drawn from the two stents, and the respective blood pressures, graft blood flow, and graft mechanical lung function are recorded.
After termination of the experiment the graft is weighed and the wet to dry ratio is determined (Sartorius Moisture Analyzer, MA30; Sartorius GmbH, Göttingen, Germany). Experiments that have failed due to technical problems are discarded from the analysis.
Experimental Groups
The correlation of overall lung function to the length of the cold ischemic interval was compared using the described measurements in five groups of rats. The groups only differed in the length of cold ischemia. The ischemic intervals were limited to 2 hours (group 1), 4 hours (group 2), 8 hours (group 3), 16 hours (group 4), and 24 hours (group 5). All lungs were preserved with reversed Euro-Collins solution and stored in the same preservation solution in a fully inflated state (26 cm H2O intratracheal pressure) at 4° to 6°C.
Statistical Analysis
Data were analyzed with the Statistical Program for Social Sciences (SPSS for Windows Version 6.0, Birmingham, UK). Continuous data were expressed as mean ± standard error. Analysis of continuous data for repeated measurements was performed as a univariate analysis of variance with Bonferroni post hoc correction of the mean. Continuous data without repeated measurements were compared by analysis of variance using the Scheffé method. Survival was analyzed as actuarial data according to the method developed by Kaplan and Meier. Graphic presentations show data as mean ± standard error of the mean.
|
Results
|
|---|
The number of experiments necessary for untrained technicians to master this model amounted to approximately 20. After this time the failure rate remained stable at 2 in 10 experiments, mainly due to technical defeats such as tearing of the pulmonary vein or artery during stent implantation. The required proficiency in microsurgery for this model is minimal as only one suture line is performed for mitral valve closure.
Reduction of artificial resistance to blood flow has been one of the major targets in model design. The internal diameter of the implanted T stents has therefore been designed larger than that of the native vessels in vivo to avoid additional resistance to blood flow. Consequently, in preliminary experiments pressures measured in the left atrium and pulmonary artery of donor and recipient are equivalent.
Quality of Donor Organ
The donor animals were very similar in terms of weight (Table 1) and quality of graft perfusion (all white, no atelectasis). The loss of static inflation pressure in donor lungs with more than 8 hours of cold ischemia amounted to not more than 2 to 4 cm H2O.
View this table:
[in this window]
[in a new window]
|
Table 1. . Comparison of Nonserial Measurements of Donor and Recipient Animal Weight, Increase in Weight of the Graft During Reperfusion, Wet to Dry Ratio, and Actuarial Survival
|
|
Organ Function After Reperfusion
During the experiment pronounced amounts of edematous fluid were suctioned from the tracheal tube, mainly in group 4 and 5 animals. These quantitative differences corresponded to changes in the weight of the donor lungs after reperfusion. There was only a little increase in weight at the end of reperfusion of the transplanted lung in groups 1, 2, and 3, whereas group 4 and 5 lungs more than doubled their weight, with the highest wet to dry ratio in group 5 animals (8.9 ± 1.9; p < 0.001). The quality of organ preservation is reflected in the length of survival. The actuarial survival was lowest in the lungs with the longest cold ischemic interval (p < 0.001).
Lung Function
The alveolar-arterial oxygen difference was lowest in the group with the shortest duration of ischemia (group 1) and increased with extension of the ischemic interval (Fig 3). The analysis of variance showed significant differences between groups 1, 2, 3 and groups 4 and 5 (p < 0.001). There was a trend toward a higher dynamic compliance in group 1, whereas both groups 4 and 5 showed a decline in the dynamic compliance, especially after the first 20 minutes of reperfusion (Fig 4). These differences, however, did not reach significance. The resistance to airflow (Fig 5) is similar in groups 1, 2, and 3. The resistance increases dramatically in the lungs of groups 4 and 5 (p < 0.04 by analysis of variance).
View larger version (28K):
[in this window]
[in a new window]
|
Fig 3. . Comparison of alveolar-arterial oxygen difference: analysis of variance with Bonferroni correction reaches significance between group 1 and groups 4 and 5 at 0, 20, 40, 60, 80, and 100 minutes (p < 0.02). The differences between group 2 and groups 4 and 5 are significant from 20 to 80 minutes after initiation of reperfusion (p < 0.01).
|
|
View larger version (40K):
[in this window]
[in a new window]
|
Fig 4. . Dynamic pulmonary compliance of the donor lung is determined as serial measurements. The compliance in group 1 is significantly higher for the entire reperfusion period when compared with group 4 and group 5 (p < 0.02).
|
|
View larger version (33K):
[in this window]
[in a new window]
|
Fig 5. . Resistance to pulmonary airflow measured in cm H2O•L-1•second-1 is significantly elevated in groups 4 and 5 when compared with groups 1, 2, and 3 (p < 0.01). The differences between group 1, group 2, and group 3 are not significant.
|
|
The hemodynamic characteristics of the graft are reflected by pulmonary vascular resistance (Fig 6). In all groups the lungs showed a markedly elevated pulmonary vascular resistance at the onset of reperfusion. Within the first 20 minutes the pulmonary vascular resistance then dropped to 200 to 300 mm Hg• ml-1• min-1 in groups 1, 2, and 3, whereas in comparison the measurements obtained from groups 4 and 5 were significantly higher (p < 0.004).
View larger version (28K):
[in this window]
[in a new window]
|
Fig 6. . Pulmonary vascular resistance is determined as the difference in pressure in pulmonary artery and vein of the donor lung divided by the graft blood flow. The pulmonary vascular resistance in group 4 and group 5 is significantly higher than group 1 and group 2 from 0 to 80 minutes after the start of reperfusion (p < 0.05).
|
|
|
Comment
|
|---|
The need for extension of the maximum tolerable ischemic interval in lung transplantation and reduction in the incidence in reperfusion injury has prompted the development of various experimental designs for evaluation of lung preservation techniques. These include basically two large groups of models using either allogenic large animals or syngeneic small rodents. Large animal models are often quite complex and expensive. The significant interindividual variability of donor and recipient animals in respect to lung function and bacterial, fungal, or viral infections are often reflected in the data obtained and limit data interpretation. To what extent large animals models more closely resemble the pathophysiology in humans when compared with experiments conducted in rodents remains questionable at this time [4, 6], and this may not warrant its preferred use over rodent models. Undoubtedly, however, large animal experiments are becoming less acceptable to the public, especially experiments conducted in dogs and primates.
The enormous amount of variables that can be altered in lung preservation require a large number of standardized experiments. Syngeneic rat models are considered ideal in this regard, but due to the small animal size and the limited blood supply these experiments often do not allow detailed and conclusive monitoring. To overcome these limitations a large number of ex situ models have been developed [7]. The ex situ models, however, do not offer the complex interaction of the lung and other organs of the recipient with the blood both as an organ and carrier of a large number of mediators at the same time. In addition, most ex situ models use mechanical pumps to simulate blood or fluid flow through the lungs with nonpulsatile flow. The parabiotic rat model uses a support animal for continuous blood supply via the vena cava for nonpulsatile perfusion of the preserved lung. The blood is returned to the support animal via a perfusion pump [8]. Besides having the disadvantage of nonpulsatile [9], low-flow perfusion, this model has a fairly large contact area between blood and foreign surfaces. The activation of the nonspecific immune system by foreign surfaces may interfere with the interactions between graft and donor [10]. In true in vivo and in situ lung transplantation models monitoring of graft performance is very limited [11] and may be reduced to a single measurement at the end of the experiment [12].
The in situ and pulsatile perfused double-lung transplantation model presented in this study seems capable of combining multiparameter monitoring with a low-cost, reproducible, and syngeneic design. This includes continuous, detailed monitoring of isolated graft function as reflected in pulmonary vascular resistance, compliance, and alveolar-arterial oxygen difference. The lungs are perfused with rat blood, and this allows for physiologic interactions between the graft and host. In comparison with the extracorporeal heart-lung transplant model [13], in which the maximum ischemic interval is limited by the ischemic tolerance of the concomitantly transplanted heart, which is necessary for perfusion of the lungs, the ischemic times of the lungs in this model are unlimited.
The results of the experiments conducted in this model show a strict correlation of the measured parameter and absolute length of cold ischemia. In group 1 with 2 hours of ischemia lungs were better preserved than in groups with longer ischemia. With a stepwise increase in the cold ischemic interval there is a decline in lung function, with the most significant deterioration occurring in the groups with longer than 8 hours of ischemia. This is quite consistent with clinical results in human transplantation, where the quality of lung function is acceptable within 8 hours of ischemia. In this experiment the reversed Euro-Collins solution was used, a solution very similar to Euro-Collins except for a low potassium concentration. Future experiments will compare the potential of different solutions to prevent or ameliorate organ damage. In addition to global lung function the actuarial survival correlates well with the quality of lung preservation. Poor lung function during the experiment is also reflected in a net gain of organ weight as well as an increase in the wet to dry weight ratio.
In the experiments conducted so far good lung function allows 25% to 40% of the total cardiac output of the recipient animal to flow through the graft. In preliminary experiments continuous clamping of the right native pulmonary artery has resulted in right ventricular failure. Therefore, continuous clamping has not been attempted in this experiment to avoid right ventricular dysfunction.
Having direct access to the pulmonary artery and vein permits sampling of blood after a single pass of drugs and mediators. The use of a double-lung block instead of a single lobe or a unilateral lung increases the area of interaction between blood and graft, decreases the vascular resistance, and increases the amount of parenchyma for bronchoalveolar lavage, histologic examination, and measurement of other physical characteristics. The strict separation of donor and recipient trachea can also be used for transtracheal application of drugs to the donor lung.
In conclusion, we believe this in situ, syngeneic rat lung transplantation model offers an excellent alternative to large animal models for verification of lung preservation solutions, for modification of donor treatment regimens, and for examining reperfusion alternatives. It is capable of combining most of the prerequisites an experimental model of acute lung transplantation should meet. Finally, this low-cost model is easily reproducible and technically easy to learn and handle.
|
Footnotes
|
|---|
Address reprint requests to Dr Hausen, Division of Thoracic and Cardiovascular Surgery, Org Nr 6210, Hannover Medical School, D-30623 Hannover, Germany.
|
References
|
|---|
- Hosenpud JD, Novick RJ, Breen TJ, Daily OP. The Registry of the International Society for Heart and Lung Transplantation: eleventh official report 1994. J Heart Lung Transplant 1994;13:561–70.[Medline]
- Cooper JD, Vreim CE. NHLBI workshop summary. Biology of lung preservation for transplantation. Am Rev Respir Dis 1992;146:803–7.[Medline]
- Colquhoun IW, Kirk AJ, Au J, et al. Single-flush perfusion with modified Euro-Collins solution: experience in clinical lung preservation. J Heart Lung Transplant 1992;11:209–14.
- Kirk AJ, Colquhoun IW, Dark JH. Lung preservation: a review of current practice and future directions. Ann Thorac Surg 1993;56:990–1000.[Abstract]
- Pickford MA, Gower JD, Dore C, Fryer PR, Green CJ. Lipid peroxidation and ultrastructural changes in rat lung isografts after single-passage organ flush and 48-hour cold storage with and without one-hour reperfusion in vivo. Transplantation 1990;50:210–8.[Medline]
- Sundaresan S, Lima O, Date H, et al. Lung preservation with low-potassium dextran flush in a primate bilateral transplant model. Ann Thorac Surg 1993;56:1129–35.[Abstract]
- Uhlig S, Wollin L. An improved setup for the isolated perfused rat lung. J Pharmacol Toxicol Methods 1994;31: 85–94.[Medline]
- Semik M, Moller F, Lange V, Bernhard A, Toomes H. Comparison of Euro-Collins and UW-1 solutions for lung preservation using the parabiotic rat perfusion model. Transplant Proc 1990;22:2235–6.[Medline]
- Meyers CH, Purut CM, D'Amico TA, Smith PK, Sabiston DCJ, Van Trigt P. Pulmonary arterial impedance after single lung transplantation. J Surg Res 1992;52:459–65.[Medline]
- Marshall DP. Ischaemic heart disease, cardiac surgery and heart/heart-lung transplantation reviewed from a haematological perspective. Br J Biomed Sci 1993;50:212–20.[Medline]
- Prop J, Wildevuur CR, Nieuwenhuis P. Acute graft-versus-host disease after lung transplantation. Transplant Proc 1989;21:2603–4.[Medline]
- Aeba R, Keenan RJ, Hardesty RL, Yousem SA, Hamamoto I, Griffith BP. University of Wisconsin solution for pulmonary preservation in a rat transplant model. Ann Thorac Surg 1992;53:240–5.[Abstract]
- Fukose T, Albes JM, Takahashi Y, Brandes H, Hausen B, Schäfers H-J. Influence of red blood cells on lung function in an ex vivo rat heart-lung model. J Surg Res 1995;59:399–404.[Medline]
This article has been cited by other articles:
|
|
|
|
 
B. Hausen, P. Mueller, M. Bahra, R. Ramsamooj, R. E. Morris, and C. W. Hewitt
Donor Treatment With the Lazeroid U74389G Reduces Ischemia-Reperfusion Injury in a Rat Lung Transplant Model
Ann. Thorac. Surg.,
September 1, 1997;
64(3):
814 - 820.
[Abstract]
[Full Text]
|
|
|
|
|
|
|
B. Hausen, R. Rohde, C. W. Hewitt, F. Schroeder, M. Beuke, R. Ramsamooj, H.-J. Schafers, Sponsor:, and H.-G. Borst
EXOGENOUS SURFACTANT TREATMENT BEFORE AND AFTER SIXTEEN HOURS OF ISCHEMIA IN EXPERIMENTAL LUNG TRANSPLANTATION
J. Thorac. Cardiovasc. Surg.,
June 1, 1997;
113(6):
1050 - 1058.
[Abstract]
[Full Text]
|
|
|
Top
Abstract
Introduction
