- Email: [email protected]
Beneficial effect of polyethylene glycol in lung preservation: early evaluation by proton nuclear magnetic resonance spectroscopy
GENERAL THORACIC
Beneficial Effect of Polyethylene Glycol in Lung Preservation: Early Evaluation by Proton Nuclear Magnetic Resonance Spectroscopy Christophe Jayle, MD, Pierre Corbi, MD, Michel Eugene, MD, PhD, Michel Carretier, MD, William Hebrard, Emmanuelle Menet, MD, and Thierry Hauet, MD, PhD Cardiothoracic Unit, Inserm ERM 324, Centre Hospitalier Universitaire de Poitiers, Poitiers; INRA, Domaine du Magneraud, Surge`res; and NMR Spectroscopy Unit, Hoˆpital Saint-Louis, Paris, France
Background. Proton nuclear magnetic resonance spectroscopy can be used to measure organic molecules in biological fluids. In this study, proton nuclear magnetic resonance spectroscopy of bronchoalveolar lavage was assessed to detect cellular damage in lung transplants. Also we evaluated a polyethylene glycol solution in lung preservation. Methods. An isolated perfused and working pig lung was used to assess initial pulmonary function after in situ cold flush and cold storage for 6 hours in three preservation solutions: (1) Euro-Collins solution, (2) University of Wisconsin solution, and (3) low potassium solution with polyethylene glycol (PEG). Pulmonary vascular resistance and partial pressure of arterial oxygen were measured during reperfusion. Bronchoalveolar lavage was studied by proton nuclear magnetic resonance spectroscopy and a histologic study of the lungs was done at the harvest after ischemia and after reperfusion. Results. Partial pressure of arterial oxygen and pulmonary vascular resistance were significantly better in PEG compared with Euro-Collins solution (p ⴝ 0.011). Inter-
stitial edema was significantly higher in Euro-Collins solution (2.4 ⴞ 0.24; p ⴝ 0.02) and University of Wisconsin solution (2.7 ⴞ 0.20; p ⴝ 0.0003) than PEG (2 ⴞ 0.16). Mitochondria scale was better in PEG (8.1 ⴞ 0.46) than in Euro-Collins solution (6.2 ⴞ 0.37; p ⴝ 0.0001) or University of Wisconsin solution (5.6 ⴞ 1.36; p ⴝ 0.0046). In bronchoalveolar lavage proton nuclear magnetic resonance spectroscopy spectra, lactate, pyruvate, citrate, and acetate were only detected after reperfusion, with a significantly reduced production of acetate in PEG. Pyruvate was reduced at the limit of significance in PEG versus University of Wisconsin solution. Conclusions. Proton nuclear magnetic resonance spectroscopy seems to be a simple and suitable method for assessment of early injury to the lung transplant. In this experimental study, PEG preserved the lung better than University of Wisconsin solution and Euro-Collins solution in both the proton nuclear magnetic resonance spectroscopy study as well as the physiologic study. (Ann Thorac Surg 2003;76:896 –902) © 2003 by The Society of Thoracic Surgeons
L
protection resides in the integrity preservation of endothelium, reduction of lipid peroxidation, and functionality improvement of pulmonary alveolar epithelium, specifically the type II alveolar pneumatocyte that synthesizes, stores and secretes pulmonary surfactant [9]. Because primary graft dysfunction is unpredictable, assessment of lung injury markers during preservation and the early postoperative period could be valuable. In recent years, clinical and experimental studies demonstrated that proton nuclear magnetic resonance spectroscopy (H-NMRS) is a powerful nondestructive and noninvasive technique that can study of a wide range of organic molecules in biological fluids in various pathologic conditions [10]. We have showed the interest of this technique in the early evaluation of renal and liver IRI after prolonged cold storage [11, 12]. Then there is a reasonable rationale to assess proton nuclear magnetic resonance spectroscopy to detect lung IRI. Hypothermic single pulmonary artery flush perfusion and cold storage allows 4 to 6 hours of ischemia before
ung transplantation has become an accepted therapeutic option for a variety of end-stage pulmonary diseases [1]. Primary graft dysfunction after lung transplantation is a significant and unpredictable complication [2, 3]. Ischemia reperfusion injury (IRI) is one of the major causes of this dysfunction and increases pulmonary transplantation morbidity and mortality [4, 5]. Pulmonary IRI results from a complex interplay of various pathogenic factors. Leukocyte-endothelial cell interaction plays a central role mediated through adhesion molecules [6, 7]. Lipid peroxidation mediated by reactive oxygen species detected in postischemic tissues promotes formation of inflammatory agents that recruit and activate neutrophils. A correlation between lipid peroxidation level in lung tissue suggested a causative role of oxidant injury in lung IRI [8]. Then the lung specificity
Accepted for publication March 25, 2003. Address reprint requests to Dr Jayle, Unite´ de Chirurgie Cardiothoracique, Pavillon Beauchant, Chu de Poitiers, 1 Rue de la Mile`trie, BP 577, 86021 Poitiers Cedex, France; e-mail: [email protected].
© 2003 by The Society of Thoracic Surgeons Published by Elsevier Inc
0003-4975/03/$30.00 S0003-4975(03)00662-3
transplantation [13]. Hyperpotassium solutions, such as Euro-Collins (EC) solution or University of Wisconsin (UW) solution are used by 90% of lung transplantation units [13]. Several studies have suggested that low potassium solution such as Celsior or Dextran could be better than standard intracellular solutions [8, 14, 15]. Interest of polyethylene glycol (PEG) has been found to improve the heart, pancreas, small bowel, kidneys, and liver preservation [16 –20]. Recently a study demonstrated that using a screening method is the ideal preservation solution for chest organs that must associate PEG and low potassium solution [21]. Mechanisms includes the prevention of osmotic swelling, decrease of lipid peroxidation [22], wash of dendritic cells, and the immunosuppressive effect [20, 23]. A previous study also demonstrated that PEG inhibited binding of several antibodies to endothelial cells and inhibited the upregulation of E-selectin [24]. The goal of this study was to test proton nuclear magnetic resonance spectroscopy of bronchoalveolar lavage (BAL) as an efficient technique to evaluate the pulmonary IRI. We also evaluated polyethylene glycol combined with a simple low potassium solution in lung preservation.
897
JAYLE ET AL NMRS FOR LUNG PRESERVATION EVALUATION
Table 1. Composition of Preservation Solutions University Polyethylene of Glycol Wisconsin Euro-Collins Solution Na⫹ (mmol/L) K⫹ (mmol/L) Mg2⫹ (mmol/L) PO43⫺ (mmol/L) SO42⫹ (mmol/L) Cl⫺ (mmol/L) HCO3⫺ (mmol/L) Ca2⫹ (mmol/L) Glucose (mmol/L) Polyethylene glycol 20000 (g/L) Pentastarch (g/L) Ac lactobionique (g/L) Raffinose (g/L) Adenosine (nmol/L) Gluthation (g/L) Allopurinol (g/L) pH a` 25°C Osmol mosm/kg
30 125 5 25 5 䡠䡠䡠 䡠䡠䡠 䡠䡠䡠 䡠䡠䡠 䡠䡠䡠 50 35.8 17.8 1.34 0.922 0.136 7.4 320
10 115 䡠䡠䡠 57.5 5 15 10 䡠䡠䡠 194 䡠䡠䡠
145 5 1.2 䡠䡠䡠 䡠䡠䡠 130 25 1.75 11 30
䡠䡠䡠 䡠䡠䡠 䡠䡠䡠 䡠䡠䡠 䡠䡠䡠 䡠䡠䡠 7.0 a´ 7.3 410
䡠䡠䡠 䡠䡠䡠 䡠䡠䡠 䡠䡠䡠 䡠䡠䡠 䡠䡠䡠 7.6
Material and Methods Fifteen Large White pigs (30 to 40 kg, INRA, Le Magneraud, Surge`res, France) were treated and cared for in accordance with the guidelines of the French agricultural office and the legislation governing animal studies. We used an isolated working and perfused pig’s left lung as a model for this study.
Animal Preparations and Surgical Procedure After intranasal injection of midazolam (0.2 mL/kg), anesthesia was induced through a conic mask by sevoflurane (8 L/mn) and maintained by propofol bolus (3 mg/kg and besides 25 mg/kg/h). An injection of adrenaline chlorhydrate lidocaine (0.5 mL/kg) was made by an epidural catheter in sacrococcygeal space. Ventilation by an endotracheal chest tube and a volume-controlled ventilator, used room air, respiratory rate (16/mn), end expiratory pressure (0 cm H2O), tidal volume to reach an ejection fraction of CO2 at 35%, and p peak inferior at 25 cm H2O. Intraoperative monitoring was carried out with an electrocardiogram and an arterial line. By a midline longitudinal sternotomy, the right and left azygos veins were ligated. The trachea was dissected. The pulmonary artery was divided and encircled with a ligature. The two venae cavae were ligated after systemic heparinization (500 U/kg), then whole blood (1 L) was sampled rapidly in a collector with anticoagulation solution (124 mL) through cannulas introduced into the inferior venae cavae. The pulmonary artery was cannulated, and the lungs were flushed by gravity at 50 cm H2O with 2 L of one of three preservation solutions at 4°C. Just after the beginning of the flush, the aorta was cross clamped and the right and the left atrium were opened. During the flush, tidal volume was reduced to obtain a peak pressure at 25 cm H2O. The topical pulmonary
cooling was facilitated by the flow of the effluent in both pleura. After flushing, the pulmonary artery was cross clamped, the trachea was ligated at the end inspiration, and the heart and lung were harvested and stored inflated in 1 L of the same flushing cold solution at 4°C. Before storage the right superior bronchus was ligated and divided to sample BAL and lung tissue.
Lung Preservation Protocol and Experimental Groups After 5 hours and 30 minutes of cold ischemia, the heart and lung block was dissected on a cold table (4°C). Only the left lung was perfused. The right pulmonary artery, the right main bronchus, the right pulmonary veins, and the median lobe veins were ligated and divided, and the right lung was removed. The left atrial was closed with a running suture around a cannula. The trachea was reopened and cannulated with a 6.5 endotracheal tube. A BAL was sampled in the medium lobe, and a lung sample was taken on the right inferior lobe. Three preservation solutions were studied: (1) EC high-potassium crystalloid solution (n ⫽ 5), (2) UW high-potassium colloid solution (n ⫽ 5), and (3) PEG solution (low potassium concentration combined with polyethylene glycol as colloid (n ⫽ 5). Composition of the three solutions are showed in Table 1. The lungs were randomly flushed and preserved in one of three solutions. Preserved lungs were compared with a control group (lungs flushed with saline solution at 4°C and just studied after the harvest [n ⫽ 5]).
Reperfusion Protocol in Isolated Perfused Worked Lung Model Whole blood of the same animal was stored 6 hours at 18°C to 24°C. A circulation of 30 minutes was made prior
GENERAL THORACIC
Ann Thorac Surg 2003;76:896 –902
898
JAYLE ET AL NMRS FOR LUNG PRESERVATION EVALUATION
GENERAL THORACIC Fig 1. Schematic diagram of the ex vivo reperfusion circuit. CDI 300 monitor, 3M (3 mol/L) (Health Care, Tustin, CA). (C stat ⫽ static compliance; PAP ⫽ pulmonary arterial pressure.)
to lung reperfusion to warming up at 38°C. Blood was diluted in ringer lactate with 19% ⫾ 2% hematocrit, deoxygenated with 2% oxygen, 5% carbon dioxide, and 93% nitrogen gas in a membrane oxygenator (Capiox SX10, Terumo Co, Tokyo, Japan) to maintain oxygen saturation of the pulmonary artery blood between 70% and 80%. Blood perfusion in the lung was done with a roller pump (Stockert-Shilley, Stockert Instrument, Mu¨ nchen, Switzerland) (Fig 1). In the reperfusion chamber, which was maintained at 38°C and humidified, perfusion gradually began at 7.5 mL/kg/min with increments of 2.5 mL/min every 5 minutes to reach systolic pulmonary artery pressure of 20 mm Hg. Then pulmonary artery flow was constantly maintained throughout the reperfusion. At the same time ventilation was gradually started at 3 mL/kg as tidal volume at 16 breaths per minute with increments of 0.5 mL/kg every 5 minutes until a tracheal peak pressure 25 cm H2O. Tidal volume was constant throughout the reperfusion. Lung perfusion time was 75 minutes. After this time lung edema increased and reperfusion was stopped.
Functional Measurements Mean pulmonary arterial pressure (mPAP in mm Hg) and reperfusion pulmonary flow (Q in mL/min) were measured every 15 minutes (RM 300, Hellige, Sweden) and allowed pulmonary vascular resistance value (pulmonary vascular resistance ⫽ mPAP/Q x 80000 in dynes/ s/cm5). Arterial and venal blood gases were measured every 15 minutes using a model CDI 400 pH/blood gas analyzer (CDI 400 monitor, 3 mol/L, Health Care, Tustin, CA). Only the partial pressure of arterial oxygen was studied for this experiment.
Morphologic and Ultrastructural Analysis At the end of the cold flush (right superior lobe), after 6 hours cold storage (right inferior lobe), and after reperfusion (left lung), lung samples were processed for light and electron microscopy. Samples for light microscopy
Ann Thorac Surg 2003;76:896 –902
study were fixed with formaldehyde 10%, embedded in paraffin, and stained with hematoxylin and eosin and periodic acid Schiff. Analysis of cellular ultrastructure using electron microscopy was performed at the same periods. Small pieces of lung tissue (1 mm3) were fixed in glutaraldehyde, washed, and postfixed in 2% osmium tetroxide for 1 hour at 4°C. After dehydration with ethanol, they were embedded in Araldite (Fluka, Buchs, Switzerland). Ultra thin sections were stained with uranyl acetate, lead citrate, and toluidine blue and then examined under an electron microscope (100CX; JOEL, Tokyo, Japan). Alveolar damages, edema and neutrophil infiltration were assessed by light microscopy. Edema was graded in 4-point scales with 0 ⫽ no edema, 1 ⫽ minor edema, 2 ⫽ mild edema, and 3 ⫽ major edema. Neutrophil infiltration was expressed as the ratio of neutrophils and leukocytes in the alveolar space, expressed in a percent. Using electron microscopy, only the type II pneumatocytes were studied, and particularly the mitochondria integrity. Mitochondria alteration was graded with a 10-point semiquantitative scale; 10 was the ideal score for normal mitochondria and 1 was the score for extremely damaged with important edema, vacuolization, and disintegration of crest.
H-NMR Spectroscopy Experiments A BAL with an infusion of saline solution (10 mL) and aspiration was collected at the end of the flush (right superior lobe), after 6 hours cold storage (medium lobe), and after reperfusion (left lung). These BAL samples were frozen at ⫺20°C until measurement was performed. Samples were thawed at room temperature and prepared with 10% D2O containing 0.75% sodium 3-[trimethyl2,2,3,3]-1-propionate (TSP) to reference the chemical shifts (0 ppm) in 5-mm nuclear magnetic resonance tubes. Spectra were acquired at 400 MHz on a Brucker AM400WB (Berlin, Germany). A water signal was suppressed with low-power irradiation (0.8 W for 0.2 seconds). Spectra acquired with 16 K data points for a spectral width of 6024 Hz with a 5-ms pulse 60° and 128 scans were accumulated. The Fourier transform was performed on 16 K data points without filtering. The resonance was identified from the literature chemical shift data or by addition of standard compounds when necessary. As line widths of the peaks were equivalent, intensity of resonance was assessed by measuring peak heights taking into account the spin-spin coupling pattern. For BAL H-NMR spectra, the level of citrate, lactate, pyruvate, and acetate were calculated and expressed in mmol/L of BAL.
Expression of Results and Statistical Analysis Results were expressed as means ⫾ standard error of the mean and were compared by using variance analysis and the Student’s t test. When a nonparametric test was needed the Kruskall-Wallis analysis was used. The exact p values for all presented results were given. Statistical analysis was done with the Number Crusher Statistical System (7.0 J.L Hintz) (1997, Paris, France).
Ann Thorac Surg 2003;76:896 –902
JAYLE ET AL NMRS FOR LUNG PRESERVATION EVALUATION
899
Group Physiological Data Pulmonary pressure of arterial oxygen (kPa)
PVR (dyn/s/cm5)
Time
Control
Euro-Collins
Polyethylene Glycol
University of Wisconsin
0⬘
6.78 ⫾ 1.01
6.20 ⫾ 2.67
4.08 ⫾ 0.66
5.14 ⫾ 0.87
15⬘ 30⬘ 45⬘ 60⬘ 75⬘ 0⬘ 15⬘ 30⬘ 45⬘ 60⬘ 75⬘
14.25 ⫾ 1.80 16.05 ⫾ 1.49 16.22 ⫾ 2.13 16.05 ⫾ 2.49 15.87 ⫾ 2.49 1,920 ⫾ 905 1,878 ⫾ 99 1,540 ⫾ 151 1,478 ⫾ 169 1,695 ⫾ 158 1,759 ⫾ 179
5.84 ⫾ 0.66 5.80 ⫾ 0.60 5.32 ⫾ 0.38 5.78 ⫾ 0.74 5.02 ⫾ 0.16 4,612 ⫾ 963 3,031 ⫾ 571 3,697 ⫾ 425 4,023 ⫾ 609 4,698 ⫾ 792 5,296 ⫾ 1,357
5.78 ⫾ 0.44 7.72 ⫾ 0.78 9.20 ⫾ 1.63 10.13 ⫾ 2.05 11.60 ⫾ 1.60 4,297 ⫾ 1,627 1,893 ⫾ 355 2,262 ⫾ 501 2,333 ⫾ 573 2,371 ⫾ 513 3,370 ⫾ 150
5.54 ⫾ 0.31 6.92 ⫾ 0.43 7.66 ⫾ 0.77 7.30 ⫾ 0.83 8.40 ⫾ 2.20 2,313 ⫾ 721 2,840 ⫾ 422 2,963 ⫾ 314 3,574 ⫾ 471 3,904 ⫾ 668 3,200 ⫾ 970
Results All the lungs were perfused for 75 minutes. After this time edema increased and reperfusion was stopped.
Effect of PEG on Partial Pressure of Arterial Oxygen and Pulmonary Vascular Resistance Measurements Partial pressure of arterial oxygen was significantly lower in the studied groups versus the control group (Table 2). However, partial pressure of arterial oxygen was significantly improved during reperfusion in PEG compared with the EC solution after 30 minutes of reperfusion (p ⫽ 0.011). There was no significant difference between the UW solution group and other experimental groups. Pulmonary vascular resistance was higher in all study groups compared with the control group (Table 2). After 30 minutes of reperfusion, pulmonary vascular resistance also was significantly lower in PEG than in the EC solution (p ⫽ 0.001), and there was no difference in the UW solution versus the EC solution or the PEG. These results suggest that PEG provides better functional results than the EC solution and is equivalent to the UW solution.
Limitation of Interstitial Edema, Neutrophils Infiltration, and Mitochondria Integrity With Partial Pressure of Arterial Oxygen Preservation Optical microscopy analysis showed an important interstitial edema after reperfusion (2.36 ⫾ 0.20), significantly greater than after harvest (0.43 ⫾ 0.10; p ⫽ 0.00001) or preservation (0.6 ⫾ 0.05; p ⫽ 0.00001). After reperfusion, interstitial edema was significantly more important in the EC solution (2.4 ⫾ 0.24; p ⫽ 0.02) and the UW solution (2.7 ⫾ 0.20; p ⫽ 0.0003) than the PEG (2 ⫾ 0.15). Associated with this edema, inflammatory infiltration was observed in preserved lungs. The ratio of neutrophils and leukocytes in the alveolar space was the same in all groups after harvest and after cold storage. After reperfusion this ratio was greater, but it was significantly smaller in the
p Value Euro-Collins vs Polyethylene Glycol 0.12 0.84 0.002 0.0001 0.002 0.0002 0.72 0.005 0.001 0.002 0.0006 0.01
PEG (44% ⫾ 8.9%) and the UW solution (40% ⫾ 7.1%) than in the EC solution (58 ⫾ 8.4; p ⫽ 0.034 and p ⫽ 0.006, respectively). There was no difference between PEG and the UW solution (p ⫽ 0.45). These results suggest a limitation of leukocyte infiltration in the interstitial space with the PEG solution and UW solution. In electronic microscopy, mitochondria of type II pneumatocytes were studied. After reperfusion, mitochondria integrity was improved in PEG (8.1 ⫾ 0.46) when compared with the EC solution (6.2 ⫾ 0.37; p ⫽ 0.0001) or UW solution (5.6 ⫾ 1.36; p ⫽ 0.0046). There was no difference between the EC solution and UW solution (p ⫽ 0.37). Mitochondria developed less edema, less vacuole, and less crest disintegration in PEG (Fig 2). Polyethylene glycol was more efficient to preserve the structure of mitochondria after reperfusion compared with the EC or UW solutions.
Evolution of H-NMR Spectroscopy Aspects and Influence of PEG In BAL, proton nuclear magnetic resonance spectroscopy, lactate, pyruvate, citrate, and acetate were not detected after lung harvest or after cold ischemia in all groups. These metabolites were significantly increased in BAL after reperfusion in all groups (Fig 3), but there was no difference in the detection of lactate and citrate after reperfusion in the different experimental groups. As shown in Table 3, there was a significantly reduced production of acetate in PEG versus the EC and UW solutions. Pyruvate excretion in BAL after reperfusion was only significantly reduced in PEG versus the UW solution (Table 3); there was no difference when compared with the EC solution.
Comment Lung ischemia reperfusion leads to increased microvascular permeability and sequestration of neutrophils in
GENERAL THORACIC
Table 2. Partial Pressure of Arterial Oxygen and Pulmonary Vascular Resistance During Reperfusion
900
JAYLE ET AL NMRS FOR LUNG PRESERVATION EVALUATION
Ann Thorac Surg 2003;76:896 –902
GENERAL THORACIC Fig 3. Proton nuclear magnetic resonance spectra of bronchoalveolar lavage after reperfusion.
Fig 2. Mitochondria type II pneumatocytes after reperfusion on electron microscopy. (EC ⫽ Euro-Collins; PEG ⫽ polyethylene glycol; UW ⫽ University of Wisconsin.)
the lungs, dysfunction of the pulmonary endothelium, and pulmonary hypertension. There is growing evidence that the immunogenic response is related to the severity of IRI [25]. Although the complex mechanisms of IRI are not yet fully understood, it is established that leukocyteendothelial cell interaction mediated through adhesion molecules is one of the key factors in IRI. In the present study, PEG solution was able to protect the lungs efficiently during cold storage and reperfusion. During normothermic reperfusion, lungs that were preserved in the PEG solution showed that the pulmonary vascular resistance significantly improved when compared with those lungs preserved in the EC solution. There was no significant difference with the UW solution. Keeping the potassium concentration to a low value reduces the entry of calcium through voltage-dependent
channels, and high sodium concentrations limit the entry of calcium through the sodium-calcium exchange. These features prevent calcium overload, avoiding a depolarization of smooth muscular cell membrane with consequential vasoconstriction and impaired distribution of the perfusion solution. At reoxygenation this reduction of vasoconstriction reduced the reperfusion injury and the swelling of endothelium when compared with high potassium solutions, which can induce swelling of the endothelium and disturb initiation of microcirculation after reperfusion [26]. As previously described, PEG is efficient as a colloid and impermeant molecule against IRI in lung preservation [27, 28]. Moreover, the unusual properties of PEG alter the interaction of cells and proteins with each other as well as with water and surfaces [24]. Polyethylene glycol conjugated to proteins shields them from interaction with other molecules, and coupling PEG to proteins or other surfaces decreases antigenicity and proteolysis [29]. These properties com-
Table 3. Concentration of Acetate and Pyruvate in Bronchoalveolar Lavage After Reperfusion
Group PEG UW EC p Value PEG vs UW PEG vs EC UW vs EC EC ⫽ Euro-Collins; of Wisconsin.
Acetate (mmol/L of Bronchoalveolar Lavage)
Pyruvate (mmol/L of Bronchoalveolar Lavage)
24.6 ⫾ 4.1 50.2 ⫾ 4.7 77.6 ⫾ 1.7
147 ⫾ 40 244 ⫾ 26 195 ⫾ 25
0.003 0.015 0.157
0.052 0.336 0.215
PEG ⫽ polyethylene glycol;
UW ⫽ University
bined with the effect of camouflaging endothelial cells support the reduction of leukocyte infiltration in the lungs preserved with the PEG solution. Proton nuclear magnetic resonance spectroscopy was able to detect lung damage compared with physiologic data or histologic analysis. In this model of an isolated, perfused, working pig lung, the proton nuclear magnetic resonance spectroscopy of BAL detected four specific metabolites after reperfusion: (1) lactate, (2) citrate, (3) acetate, and (4) pyruvate. Lactate and citrate are not fully efficient to give information in this study because the oxygenation membrane of the pump was washed with lactate solution and exogenous citrate was used for samples and blood conservation; further studies are necessary for these data. However, acetate and pyruvate are more discriminated in these models. Acetate concentration after reperfusion was significantly lower in the PEG than with the EC or UW solutions. There are no differences between EC and the UW solutions. We and others have previously detected that acetate was increased after IRI and that it was related to organ dysfunction [10, 30]. Because acetate is the main acetyl coenzyme A provider for the citrate synthase reaction, which is the first step of the citric acid cycle, the increase of acetate could be related with mitochondria dysfunction and anaerobic metabolism conditions. The significantly lower acetate detection in PEG was a marker of a better mitochondria metabolism than in the UW or EC solutions. Pyruvate was also detected in BAL after reperfusion in all groups. The rate was significantly lower in PEG than in the UW solution, and there was no difference between the UW and EC solutions, and no difference between the PEG and EC solutions. Pyruvate is transported in the mitochondria, and after an oxidative decarboxylation, it is transformed into acetyl coenzyme A and used for the citric acid cycle. There could be a relationship between the decrease of pyruvate and acetate in PEG after reperfusion and a better mitochondrial preservation in this group than in the EC and UW solutions. These results were correlated with the score of mitochondria after reperfusion in PEG in electronic microscopy analysis. The mitochondria integrity after reperfusion was improved in PEG. Lung specificity resides in necessity to preserve the integrity and functionality of the pulmonary alveolar epithelium, especially the type II alveolar pneumatocyte that synthesizes, stores, and secretes pulmonary surfactant [9]. In conclusion, the use of an extracellular solution combined with PEG, as an impermeant in lung conservation, could be useful for as long as 6 hours of cold storage. This study also demonstrated the potential interest of proton nuclear magnetic resonance spectroscopy in the evaluation of lungs in the early phase of reperfusion. Proton nuclear magnetic resonance spectroscopy gives information on mitochondria status and the level of reperfusion damage. This method could be interesting to assess grafts during the early period of reperfusion.
JAYLE ET AL NMRS FOR LUNG PRESERVATION EVALUATION
901
This work was supported by grants from the Fondation Pour la Recherche Me´ dicale, Paris, France. We thank Catherine Henry, Raymond Poncharaud, Nathalie Quellard, and Beatrice Fernandez for their technical assistance. We thank the Society Terumo for the donation of Oxygenator Capiox SX10 (Terumo Corp, Tokyo, Japan).
References 1. Cooper JD. Lung transplantation—a new era. Ann Thorac Surg 1987;44:447–8. 2. Christie JD, Bavaria JE, Palesky HI, et al. Primary graft failure following lung transplantation. Chest 1998;114:51–60. 3. Schulman LL. Perioperative mortality and primary graft failure. Chest 1998;114:7–8. 4. Hosenpud JD, Novick RJ, Bennet LE, et al. The registry of the International Society for Heart and Lung transplantation: thirteenth official report, 1996. J Heart Lung Transpl 1996;15: 655. 5. Breda MA, Hall TS, Stuart S, et al. Twenty-four hour lung preservation by hypothermia, and leukocyte depletion. J Heart Transplant 1985;4:325–9. 6. Novick RJ, Gehman KE, Ali IS, Lee J. Lung preservation: the importance of endothelium and alveolar type II cell integrity. Ann Thorac Surg 1996;62:302–14. 7. Lefer AM, Lefer DJ. Pharmacology of the endothelium in ischemia-reperfusion and circulatory shock. Ann Rev Pharmacol Toxicol 1993;33:71–90. 8. Sakamaki F, Hoffmann H, Muller C, Dienemann H, Messmer K, Schildberg FW. Reduced lipid peroxidation and ischemia reperfusion injury after lung transplantation using low potassium dextran solution for lung preservation. Am J Respir Crit Care Med 1997;156:1073–81. 9. Novick RJ, Possmayer F, Veldhuizen RAW, et al. Surfactant analysis and replacement therapy: a future tool of the lung transplant surgeon? Ann Thorac Surg 1991;52:1194 –200. 10. Neild GH, Foxall PJD, Lindon JC, Holmes EC, Nicholson JK. Uroscopy in the 21st century: high field NMR spectroscopy. Nephrol Dial Transpl 1997;12:404 –17. 11. Gibelin H, Eugene M, Hebrard W, Henry C, Carretier M, Hauet T. A new approach to the evaluation of liver graft function by nuclear magnetic resonance spectroscopy. A comparative study between Euro-Collins and University of Wisconsin solutions. Clin Chem Lab Med 2000;38:1133–6. 12. Hauet T, Goujon JM, Tallineau C, Carretier M, Euge`ne M. Early evaluation of renal reperfusion injury after prolonged cold storage using proton nuclear magnetic resonance spectroscopy. Br J Surg 1999;86:401–9. 13. Hopkinson DN, Bhabra MS, Hooper TL. Pulmonary graft preservation: a worldwide survey of current critical practice. J Heart Lung Transpl 1998;17:525–31. 14. Xiong L, Legagneux J, Wassef M, et al. Protective effects of Celsior in lung transplantation. J Heart Lung Transpl 1999; 18:320 –7. 15. Moriyasu K, McKeown PP, Novitzky D, Snow TR. Preservation of competent rabbit lung function after 30 hours of storage with low-potassium dextran solution. J Heart Lung Transpl 1995;14:75–9. 16. Wicomb WN, Hill JD, Avery J, Collins GM. Optimal cardioplegia and 24-hour heart storage with simplified UW containing polyethylene glycol. Transplantation 1990;49:261. 17. Zheng T, Lanza RP, Soon-Shiong P. Prolonged pancreas preservation using a simplified solution containing polyethylene glycol. Transplantation 1991;51:63–6. 18. Itasaka H, Burns W, Wicomb WN, Egaya H, Collins G, Esquivel CO. Modification of rejection by polyethylene glycol in small bowel transplantation. Transplantation 1994;57: 645–8. 19. Hauet T, Baumert H, Amor I, et al. Protection of autotransplantated pig kidneys from ischemia-reperfusion injury by polyethylene glycol. Transplantation 2000;70:1569 –75.
GENERAL THORACIC
Ann Thorac Surg 2003;76:896 –902
902
JAYLE ET AL NMRS FOR LUNG PRESERVATION EVALUATION
GENERAL THORACIC
20. Tokunaga Y, Wicomb WN, Garcia-Kennedy R, Esquivel CO, Collins GM. The immunosuppressive effect of polyethylene glycol in a flush solution for rat liver transplantation. Transplantation 1992;54:756 –8. 21. Ferrera R, Michel P, Hadour G, Rodriguez C, Ovize M, Phan Tan Luu R. An optimal experimental design for the development of preservative heart solutions. J Heart Lung Transpl 2002;21:260 –70. 22. Hauet T, Baumert H, Faure JP, et al. Beneficial effects of low-potassium and polyethylene glycol solution on renal lipid peroxidation during 48-hour cold storage and normothermic reperfusion. Transplant Proc 1998;30:2798 –9. 23. Collins GM, Wicomb WN, Levin BS, Verma S, Avery J, Hill JD. Heart preservation solution containing polyethylene glycol: an immunosuppressive effect? Lancet 1991;338:890 –1. 24. Sthulmetier KM, Lin Y. Camouflaging endothelial cells: does it prolong graft survival? Bioch Bioph Acta 1999;1428:177–90. 25. Qayami AK, Nikbakht-Sangari M, Godin DV, et al. The relationship of ischemia-reperfusion injury of transplanted
Ann Thorac Surg 2003;76:896 –902
26. 27.
28.
29. 30.
lung and the up-regulation of major histocompatibility complex II on host peripheral. J Thorac Cardiovasc Surg 1998; 115:978 –89. Wheeldon D, Sharples L, Wallwork J, English T. Donor heart preservation survey. J Heart Lung Transpl 1992;11:986 –93. Eugene M, Hauet T, Mothes D, et al. Beneficial effects of a low-potassium and polyethylene glycol solution on renal function and structure during 48-hour cold storage preservation. Transplant Proc 1997;29:2360 –2. Hauet T, Mothes D, Goujon JM, Carretier M, Eugene M. Protective effect of polyethylene glycol against prolonged cold ischemia and reperfusion injury: study in the isolated perfused rat kidney. J Pharmacol Exp Ther 2001;297:946 –52. Park KD, Kim YS, Han DK, et al. Bacterial adhesion on PEG modified polyurethane surfaces. Biomaterials 1998;19:851–9. Hauet T, Mothes D, Bon D, et al. Proton NMR spectroscopy as a novel approach to the monitoring of citrate, and trimethylamine-N-oxyde excretion after kidney preservation. Transplant Proc 1997;29:2323–5.
Beneficial Effect of Polyethylene Glycol in Lung Preservation: Early Evaluation by Proton Nuclear Magnetic Resonance Spectroscopy Christophe Jayle, MD, Pierre Corbi, MD, Michel Eugene, MD, PhD, Michel Carretier, MD, William Hebrard, Emmanuelle Menet, MD, and Thierry Hauet, MD, PhD Cardiothoracic Unit, Inserm ERM 324, Centre Hospitalier Universitaire de Poitiers, Poitiers; INRA, Domaine du Magneraud, Surge`res; and NMR Spectroscopy Unit, Hoˆpital Saint-Louis, Paris, France
Background. Proton nuclear magnetic resonance spectroscopy can be used to measure organic molecules in biological fluids. In this study, proton nuclear magnetic resonance spectroscopy of bronchoalveolar lavage was assessed to detect cellular damage in lung transplants. Also we evaluated a polyethylene glycol solution in lung preservation. Methods. An isolated perfused and working pig lung was used to assess initial pulmonary function after in situ cold flush and cold storage for 6 hours in three preservation solutions: (1) Euro-Collins solution, (2) University of Wisconsin solution, and (3) low potassium solution with polyethylene glycol (PEG). Pulmonary vascular resistance and partial pressure of arterial oxygen were measured during reperfusion. Bronchoalveolar lavage was studied by proton nuclear magnetic resonance spectroscopy and a histologic study of the lungs was done at the harvest after ischemia and after reperfusion. Results. Partial pressure of arterial oxygen and pulmonary vascular resistance were significantly better in PEG compared with Euro-Collins solution (p ⴝ 0.011). Inter-
stitial edema was significantly higher in Euro-Collins solution (2.4 ⴞ 0.24; p ⴝ 0.02) and University of Wisconsin solution (2.7 ⴞ 0.20; p ⴝ 0.0003) than PEG (2 ⴞ 0.16). Mitochondria scale was better in PEG (8.1 ⴞ 0.46) than in Euro-Collins solution (6.2 ⴞ 0.37; p ⴝ 0.0001) or University of Wisconsin solution (5.6 ⴞ 1.36; p ⴝ 0.0046). In bronchoalveolar lavage proton nuclear magnetic resonance spectroscopy spectra, lactate, pyruvate, citrate, and acetate were only detected after reperfusion, with a significantly reduced production of acetate in PEG. Pyruvate was reduced at the limit of significance in PEG versus University of Wisconsin solution. Conclusions. Proton nuclear magnetic resonance spectroscopy seems to be a simple and suitable method for assessment of early injury to the lung transplant. In this experimental study, PEG preserved the lung better than University of Wisconsin solution and Euro-Collins solution in both the proton nuclear magnetic resonance spectroscopy study as well as the physiologic study. (Ann Thorac Surg 2003;76:896 –902) © 2003 by The Society of Thoracic Surgeons
L
protection resides in the integrity preservation of endothelium, reduction of lipid peroxidation, and functionality improvement of pulmonary alveolar epithelium, specifically the type II alveolar pneumatocyte that synthesizes, stores and secretes pulmonary surfactant [9]. Because primary graft dysfunction is unpredictable, assessment of lung injury markers during preservation and the early postoperative period could be valuable. In recent years, clinical and experimental studies demonstrated that proton nuclear magnetic resonance spectroscopy (H-NMRS) is a powerful nondestructive and noninvasive technique that can study of a wide range of organic molecules in biological fluids in various pathologic conditions [10]. We have showed the interest of this technique in the early evaluation of renal and liver IRI after prolonged cold storage [11, 12]. Then there is a reasonable rationale to assess proton nuclear magnetic resonance spectroscopy to detect lung IRI. Hypothermic single pulmonary artery flush perfusion and cold storage allows 4 to 6 hours of ischemia before
ung transplantation has become an accepted therapeutic option for a variety of end-stage pulmonary diseases [1]. Primary graft dysfunction after lung transplantation is a significant and unpredictable complication [2, 3]. Ischemia reperfusion injury (IRI) is one of the major causes of this dysfunction and increases pulmonary transplantation morbidity and mortality [4, 5]. Pulmonary IRI results from a complex interplay of various pathogenic factors. Leukocyte-endothelial cell interaction plays a central role mediated through adhesion molecules [6, 7]. Lipid peroxidation mediated by reactive oxygen species detected in postischemic tissues promotes formation of inflammatory agents that recruit and activate neutrophils. A correlation between lipid peroxidation level in lung tissue suggested a causative role of oxidant injury in lung IRI [8]. Then the lung specificity
Accepted for publication March 25, 2003. Address reprint requests to Dr Jayle, Unite´ de Chirurgie Cardiothoracique, Pavillon Beauchant, Chu de Poitiers, 1 Rue de la Mile`trie, BP 577, 86021 Poitiers Cedex, France; e-mail: [email protected].
© 2003 by The Society of Thoracic Surgeons Published by Elsevier Inc
0003-4975/03/$30.00 S0003-4975(03)00662-3
transplantation [13]. Hyperpotassium solutions, such as Euro-Collins (EC) solution or University of Wisconsin (UW) solution are used by 90% of lung transplantation units [13]. Several studies have suggested that low potassium solution such as Celsior or Dextran could be better than standard intracellular solutions [8, 14, 15]. Interest of polyethylene glycol (PEG) has been found to improve the heart, pancreas, small bowel, kidneys, and liver preservation [16 –20]. Recently a study demonstrated that using a screening method is the ideal preservation solution for chest organs that must associate PEG and low potassium solution [21]. Mechanisms includes the prevention of osmotic swelling, decrease of lipid peroxidation [22], wash of dendritic cells, and the immunosuppressive effect [20, 23]. A previous study also demonstrated that PEG inhibited binding of several antibodies to endothelial cells and inhibited the upregulation of E-selectin [24]. The goal of this study was to test proton nuclear magnetic resonance spectroscopy of bronchoalveolar lavage (BAL) as an efficient technique to evaluate the pulmonary IRI. We also evaluated polyethylene glycol combined with a simple low potassium solution in lung preservation.
897
JAYLE ET AL NMRS FOR LUNG PRESERVATION EVALUATION
Table 1. Composition of Preservation Solutions University Polyethylene of Glycol Wisconsin Euro-Collins Solution Na⫹ (mmol/L) K⫹ (mmol/L) Mg2⫹ (mmol/L) PO43⫺ (mmol/L) SO42⫹ (mmol/L) Cl⫺ (mmol/L) HCO3⫺ (mmol/L) Ca2⫹ (mmol/L) Glucose (mmol/L) Polyethylene glycol 20000 (g/L) Pentastarch (g/L) Ac lactobionique (g/L) Raffinose (g/L) Adenosine (nmol/L) Gluthation (g/L) Allopurinol (g/L) pH a` 25°C Osmol mosm/kg
30 125 5 25 5 䡠䡠䡠 䡠䡠䡠 䡠䡠䡠 䡠䡠䡠 䡠䡠䡠 50 35.8 17.8 1.34 0.922 0.136 7.4 320
10 115 䡠䡠䡠 57.5 5 15 10 䡠䡠䡠 194 䡠䡠䡠
145 5 1.2 䡠䡠䡠 䡠䡠䡠 130 25 1.75 11 30
䡠䡠䡠 䡠䡠䡠 䡠䡠䡠 䡠䡠䡠 䡠䡠䡠 䡠䡠䡠 7.0 a´ 7.3 410
䡠䡠䡠 䡠䡠䡠 䡠䡠䡠 䡠䡠䡠 䡠䡠䡠 䡠䡠䡠 7.6
Material and Methods Fifteen Large White pigs (30 to 40 kg, INRA, Le Magneraud, Surge`res, France) were treated and cared for in accordance with the guidelines of the French agricultural office and the legislation governing animal studies. We used an isolated working and perfused pig’s left lung as a model for this study.
Animal Preparations and Surgical Procedure After intranasal injection of midazolam (0.2 mL/kg), anesthesia was induced through a conic mask by sevoflurane (8 L/mn) and maintained by propofol bolus (3 mg/kg and besides 25 mg/kg/h). An injection of adrenaline chlorhydrate lidocaine (0.5 mL/kg) was made by an epidural catheter in sacrococcygeal space. Ventilation by an endotracheal chest tube and a volume-controlled ventilator, used room air, respiratory rate (16/mn), end expiratory pressure (0 cm H2O), tidal volume to reach an ejection fraction of CO2 at 35%, and p peak inferior at 25 cm H2O. Intraoperative monitoring was carried out with an electrocardiogram and an arterial line. By a midline longitudinal sternotomy, the right and left azygos veins were ligated. The trachea was dissected. The pulmonary artery was divided and encircled with a ligature. The two venae cavae were ligated after systemic heparinization (500 U/kg), then whole blood (1 L) was sampled rapidly in a collector with anticoagulation solution (124 mL) through cannulas introduced into the inferior venae cavae. The pulmonary artery was cannulated, and the lungs were flushed by gravity at 50 cm H2O with 2 L of one of three preservation solutions at 4°C. Just after the beginning of the flush, the aorta was cross clamped and the right and the left atrium were opened. During the flush, tidal volume was reduced to obtain a peak pressure at 25 cm H2O. The topical pulmonary
cooling was facilitated by the flow of the effluent in both pleura. After flushing, the pulmonary artery was cross clamped, the trachea was ligated at the end inspiration, and the heart and lung were harvested and stored inflated in 1 L of the same flushing cold solution at 4°C. Before storage the right superior bronchus was ligated and divided to sample BAL and lung tissue.
Lung Preservation Protocol and Experimental Groups After 5 hours and 30 minutes of cold ischemia, the heart and lung block was dissected on a cold table (4°C). Only the left lung was perfused. The right pulmonary artery, the right main bronchus, the right pulmonary veins, and the median lobe veins were ligated and divided, and the right lung was removed. The left atrial was closed with a running suture around a cannula. The trachea was reopened and cannulated with a 6.5 endotracheal tube. A BAL was sampled in the medium lobe, and a lung sample was taken on the right inferior lobe. Three preservation solutions were studied: (1) EC high-potassium crystalloid solution (n ⫽ 5), (2) UW high-potassium colloid solution (n ⫽ 5), and (3) PEG solution (low potassium concentration combined with polyethylene glycol as colloid (n ⫽ 5). Composition of the three solutions are showed in Table 1. The lungs were randomly flushed and preserved in one of three solutions. Preserved lungs were compared with a control group (lungs flushed with saline solution at 4°C and just studied after the harvest [n ⫽ 5]).
Reperfusion Protocol in Isolated Perfused Worked Lung Model Whole blood of the same animal was stored 6 hours at 18°C to 24°C. A circulation of 30 minutes was made prior
GENERAL THORACIC
Ann Thorac Surg 2003;76:896 –902
898
JAYLE ET AL NMRS FOR LUNG PRESERVATION EVALUATION
GENERAL THORACIC Fig 1. Schematic diagram of the ex vivo reperfusion circuit. CDI 300 monitor, 3M (3 mol/L) (Health Care, Tustin, CA). (C stat ⫽ static compliance; PAP ⫽ pulmonary arterial pressure.)
to lung reperfusion to warming up at 38°C. Blood was diluted in ringer lactate with 19% ⫾ 2% hematocrit, deoxygenated with 2% oxygen, 5% carbon dioxide, and 93% nitrogen gas in a membrane oxygenator (Capiox SX10, Terumo Co, Tokyo, Japan) to maintain oxygen saturation of the pulmonary artery blood between 70% and 80%. Blood perfusion in the lung was done with a roller pump (Stockert-Shilley, Stockert Instrument, Mu¨ nchen, Switzerland) (Fig 1). In the reperfusion chamber, which was maintained at 38°C and humidified, perfusion gradually began at 7.5 mL/kg/min with increments of 2.5 mL/min every 5 minutes to reach systolic pulmonary artery pressure of 20 mm Hg. Then pulmonary artery flow was constantly maintained throughout the reperfusion. At the same time ventilation was gradually started at 3 mL/kg as tidal volume at 16 breaths per minute with increments of 0.5 mL/kg every 5 minutes until a tracheal peak pressure 25 cm H2O. Tidal volume was constant throughout the reperfusion. Lung perfusion time was 75 minutes. After this time lung edema increased and reperfusion was stopped.
Functional Measurements Mean pulmonary arterial pressure (mPAP in mm Hg) and reperfusion pulmonary flow (Q in mL/min) were measured every 15 minutes (RM 300, Hellige, Sweden) and allowed pulmonary vascular resistance value (pulmonary vascular resistance ⫽ mPAP/Q x 80000 in dynes/ s/cm5). Arterial and venal blood gases were measured every 15 minutes using a model CDI 400 pH/blood gas analyzer (CDI 400 monitor, 3 mol/L, Health Care, Tustin, CA). Only the partial pressure of arterial oxygen was studied for this experiment.
Morphologic and Ultrastructural Analysis At the end of the cold flush (right superior lobe), after 6 hours cold storage (right inferior lobe), and after reperfusion (left lung), lung samples were processed for light and electron microscopy. Samples for light microscopy
Ann Thorac Surg 2003;76:896 –902
study were fixed with formaldehyde 10%, embedded in paraffin, and stained with hematoxylin and eosin and periodic acid Schiff. Analysis of cellular ultrastructure using electron microscopy was performed at the same periods. Small pieces of lung tissue (1 mm3) were fixed in glutaraldehyde, washed, and postfixed in 2% osmium tetroxide for 1 hour at 4°C. After dehydration with ethanol, they were embedded in Araldite (Fluka, Buchs, Switzerland). Ultra thin sections were stained with uranyl acetate, lead citrate, and toluidine blue and then examined under an electron microscope (100CX; JOEL, Tokyo, Japan). Alveolar damages, edema and neutrophil infiltration were assessed by light microscopy. Edema was graded in 4-point scales with 0 ⫽ no edema, 1 ⫽ minor edema, 2 ⫽ mild edema, and 3 ⫽ major edema. Neutrophil infiltration was expressed as the ratio of neutrophils and leukocytes in the alveolar space, expressed in a percent. Using electron microscopy, only the type II pneumatocytes were studied, and particularly the mitochondria integrity. Mitochondria alteration was graded with a 10-point semiquantitative scale; 10 was the ideal score for normal mitochondria and 1 was the score for extremely damaged with important edema, vacuolization, and disintegration of crest.
H-NMR Spectroscopy Experiments A BAL with an infusion of saline solution (10 mL) and aspiration was collected at the end of the flush (right superior lobe), after 6 hours cold storage (medium lobe), and after reperfusion (left lung). These BAL samples were frozen at ⫺20°C until measurement was performed. Samples were thawed at room temperature and prepared with 10% D2O containing 0.75% sodium 3-[trimethyl2,2,3,3]-1-propionate (TSP) to reference the chemical shifts (0 ppm) in 5-mm nuclear magnetic resonance tubes. Spectra were acquired at 400 MHz on a Brucker AM400WB (Berlin, Germany). A water signal was suppressed with low-power irradiation (0.8 W for 0.2 seconds). Spectra acquired with 16 K data points for a spectral width of 6024 Hz with a 5-ms pulse 60° and 128 scans were accumulated. The Fourier transform was performed on 16 K data points without filtering. The resonance was identified from the literature chemical shift data or by addition of standard compounds when necessary. As line widths of the peaks were equivalent, intensity of resonance was assessed by measuring peak heights taking into account the spin-spin coupling pattern. For BAL H-NMR spectra, the level of citrate, lactate, pyruvate, and acetate were calculated and expressed in mmol/L of BAL.
Expression of Results and Statistical Analysis Results were expressed as means ⫾ standard error of the mean and were compared by using variance analysis and the Student’s t test. When a nonparametric test was needed the Kruskall-Wallis analysis was used. The exact p values for all presented results were given. Statistical analysis was done with the Number Crusher Statistical System (7.0 J.L Hintz) (1997, Paris, France).
Ann Thorac Surg 2003;76:896 –902
JAYLE ET AL NMRS FOR LUNG PRESERVATION EVALUATION
899
Group Physiological Data Pulmonary pressure of arterial oxygen (kPa)
PVR (dyn/s/cm5)
Time
Control
Euro-Collins
Polyethylene Glycol
University of Wisconsin
0⬘
6.78 ⫾ 1.01
6.20 ⫾ 2.67
4.08 ⫾ 0.66
5.14 ⫾ 0.87
15⬘ 30⬘ 45⬘ 60⬘ 75⬘ 0⬘ 15⬘ 30⬘ 45⬘ 60⬘ 75⬘
14.25 ⫾ 1.80 16.05 ⫾ 1.49 16.22 ⫾ 2.13 16.05 ⫾ 2.49 15.87 ⫾ 2.49 1,920 ⫾ 905 1,878 ⫾ 99 1,540 ⫾ 151 1,478 ⫾ 169 1,695 ⫾ 158 1,759 ⫾ 179
5.84 ⫾ 0.66 5.80 ⫾ 0.60 5.32 ⫾ 0.38 5.78 ⫾ 0.74 5.02 ⫾ 0.16 4,612 ⫾ 963 3,031 ⫾ 571 3,697 ⫾ 425 4,023 ⫾ 609 4,698 ⫾ 792 5,296 ⫾ 1,357
5.78 ⫾ 0.44 7.72 ⫾ 0.78 9.20 ⫾ 1.63 10.13 ⫾ 2.05 11.60 ⫾ 1.60 4,297 ⫾ 1,627 1,893 ⫾ 355 2,262 ⫾ 501 2,333 ⫾ 573 2,371 ⫾ 513 3,370 ⫾ 150
5.54 ⫾ 0.31 6.92 ⫾ 0.43 7.66 ⫾ 0.77 7.30 ⫾ 0.83 8.40 ⫾ 2.20 2,313 ⫾ 721 2,840 ⫾ 422 2,963 ⫾ 314 3,574 ⫾ 471 3,904 ⫾ 668 3,200 ⫾ 970
Results All the lungs were perfused for 75 minutes. After this time edema increased and reperfusion was stopped.
Effect of PEG on Partial Pressure of Arterial Oxygen and Pulmonary Vascular Resistance Measurements Partial pressure of arterial oxygen was significantly lower in the studied groups versus the control group (Table 2). However, partial pressure of arterial oxygen was significantly improved during reperfusion in PEG compared with the EC solution after 30 minutes of reperfusion (p ⫽ 0.011). There was no significant difference between the UW solution group and other experimental groups. Pulmonary vascular resistance was higher in all study groups compared with the control group (Table 2). After 30 minutes of reperfusion, pulmonary vascular resistance also was significantly lower in PEG than in the EC solution (p ⫽ 0.001), and there was no difference in the UW solution versus the EC solution or the PEG. These results suggest that PEG provides better functional results than the EC solution and is equivalent to the UW solution.
Limitation of Interstitial Edema, Neutrophils Infiltration, and Mitochondria Integrity With Partial Pressure of Arterial Oxygen Preservation Optical microscopy analysis showed an important interstitial edema after reperfusion (2.36 ⫾ 0.20), significantly greater than after harvest (0.43 ⫾ 0.10; p ⫽ 0.00001) or preservation (0.6 ⫾ 0.05; p ⫽ 0.00001). After reperfusion, interstitial edema was significantly more important in the EC solution (2.4 ⫾ 0.24; p ⫽ 0.02) and the UW solution (2.7 ⫾ 0.20; p ⫽ 0.0003) than the PEG (2 ⫾ 0.15). Associated with this edema, inflammatory infiltration was observed in preserved lungs. The ratio of neutrophils and leukocytes in the alveolar space was the same in all groups after harvest and after cold storage. After reperfusion this ratio was greater, but it was significantly smaller in the
p Value Euro-Collins vs Polyethylene Glycol 0.12 0.84 0.002 0.0001 0.002 0.0002 0.72 0.005 0.001 0.002 0.0006 0.01
PEG (44% ⫾ 8.9%) and the UW solution (40% ⫾ 7.1%) than in the EC solution (58 ⫾ 8.4; p ⫽ 0.034 and p ⫽ 0.006, respectively). There was no difference between PEG and the UW solution (p ⫽ 0.45). These results suggest a limitation of leukocyte infiltration in the interstitial space with the PEG solution and UW solution. In electronic microscopy, mitochondria of type II pneumatocytes were studied. After reperfusion, mitochondria integrity was improved in PEG (8.1 ⫾ 0.46) when compared with the EC solution (6.2 ⫾ 0.37; p ⫽ 0.0001) or UW solution (5.6 ⫾ 1.36; p ⫽ 0.0046). There was no difference between the EC solution and UW solution (p ⫽ 0.37). Mitochondria developed less edema, less vacuole, and less crest disintegration in PEG (Fig 2). Polyethylene glycol was more efficient to preserve the structure of mitochondria after reperfusion compared with the EC or UW solutions.
Evolution of H-NMR Spectroscopy Aspects and Influence of PEG In BAL, proton nuclear magnetic resonance spectroscopy, lactate, pyruvate, citrate, and acetate were not detected after lung harvest or after cold ischemia in all groups. These metabolites were significantly increased in BAL after reperfusion in all groups (Fig 3), but there was no difference in the detection of lactate and citrate after reperfusion in the different experimental groups. As shown in Table 3, there was a significantly reduced production of acetate in PEG versus the EC and UW solutions. Pyruvate excretion in BAL after reperfusion was only significantly reduced in PEG versus the UW solution (Table 3); there was no difference when compared with the EC solution.
Comment Lung ischemia reperfusion leads to increased microvascular permeability and sequestration of neutrophils in
GENERAL THORACIC
Table 2. Partial Pressure of Arterial Oxygen and Pulmonary Vascular Resistance During Reperfusion
900
JAYLE ET AL NMRS FOR LUNG PRESERVATION EVALUATION
Ann Thorac Surg 2003;76:896 –902
GENERAL THORACIC Fig 3. Proton nuclear magnetic resonance spectra of bronchoalveolar lavage after reperfusion.
Fig 2. Mitochondria type II pneumatocytes after reperfusion on electron microscopy. (EC ⫽ Euro-Collins; PEG ⫽ polyethylene glycol; UW ⫽ University of Wisconsin.)
the lungs, dysfunction of the pulmonary endothelium, and pulmonary hypertension. There is growing evidence that the immunogenic response is related to the severity of IRI [25]. Although the complex mechanisms of IRI are not yet fully understood, it is established that leukocyteendothelial cell interaction mediated through adhesion molecules is one of the key factors in IRI. In the present study, PEG solution was able to protect the lungs efficiently during cold storage and reperfusion. During normothermic reperfusion, lungs that were preserved in the PEG solution showed that the pulmonary vascular resistance significantly improved when compared with those lungs preserved in the EC solution. There was no significant difference with the UW solution. Keeping the potassium concentration to a low value reduces the entry of calcium through voltage-dependent
channels, and high sodium concentrations limit the entry of calcium through the sodium-calcium exchange. These features prevent calcium overload, avoiding a depolarization of smooth muscular cell membrane with consequential vasoconstriction and impaired distribution of the perfusion solution. At reoxygenation this reduction of vasoconstriction reduced the reperfusion injury and the swelling of endothelium when compared with high potassium solutions, which can induce swelling of the endothelium and disturb initiation of microcirculation after reperfusion [26]. As previously described, PEG is efficient as a colloid and impermeant molecule against IRI in lung preservation [27, 28]. Moreover, the unusual properties of PEG alter the interaction of cells and proteins with each other as well as with water and surfaces [24]. Polyethylene glycol conjugated to proteins shields them from interaction with other molecules, and coupling PEG to proteins or other surfaces decreases antigenicity and proteolysis [29]. These properties com-
Table 3. Concentration of Acetate and Pyruvate in Bronchoalveolar Lavage After Reperfusion
Group PEG UW EC p Value PEG vs UW PEG vs EC UW vs EC EC ⫽ Euro-Collins; of Wisconsin.
Acetate (mmol/L of Bronchoalveolar Lavage)
Pyruvate (mmol/L of Bronchoalveolar Lavage)
24.6 ⫾ 4.1 50.2 ⫾ 4.7 77.6 ⫾ 1.7
147 ⫾ 40 244 ⫾ 26 195 ⫾ 25
0.003 0.015 0.157
0.052 0.336 0.215
PEG ⫽ polyethylene glycol;
UW ⫽ University
bined with the effect of camouflaging endothelial cells support the reduction of leukocyte infiltration in the lungs preserved with the PEG solution. Proton nuclear magnetic resonance spectroscopy was able to detect lung damage compared with physiologic data or histologic analysis. In this model of an isolated, perfused, working pig lung, the proton nuclear magnetic resonance spectroscopy of BAL detected four specific metabolites after reperfusion: (1) lactate, (2) citrate, (3) acetate, and (4) pyruvate. Lactate and citrate are not fully efficient to give information in this study because the oxygenation membrane of the pump was washed with lactate solution and exogenous citrate was used for samples and blood conservation; further studies are necessary for these data. However, acetate and pyruvate are more discriminated in these models. Acetate concentration after reperfusion was significantly lower in the PEG than with the EC or UW solutions. There are no differences between EC and the UW solutions. We and others have previously detected that acetate was increased after IRI and that it was related to organ dysfunction [10, 30]. Because acetate is the main acetyl coenzyme A provider for the citrate synthase reaction, which is the first step of the citric acid cycle, the increase of acetate could be related with mitochondria dysfunction and anaerobic metabolism conditions. The significantly lower acetate detection in PEG was a marker of a better mitochondria metabolism than in the UW or EC solutions. Pyruvate was also detected in BAL after reperfusion in all groups. The rate was significantly lower in PEG than in the UW solution, and there was no difference between the UW and EC solutions, and no difference between the PEG and EC solutions. Pyruvate is transported in the mitochondria, and after an oxidative decarboxylation, it is transformed into acetyl coenzyme A and used for the citric acid cycle. There could be a relationship between the decrease of pyruvate and acetate in PEG after reperfusion and a better mitochondrial preservation in this group than in the EC and UW solutions. These results were correlated with the score of mitochondria after reperfusion in PEG in electronic microscopy analysis. The mitochondria integrity after reperfusion was improved in PEG. Lung specificity resides in necessity to preserve the integrity and functionality of the pulmonary alveolar epithelium, especially the type II alveolar pneumatocyte that synthesizes, stores, and secretes pulmonary surfactant [9]. In conclusion, the use of an extracellular solution combined with PEG, as an impermeant in lung conservation, could be useful for as long as 6 hours of cold storage. This study also demonstrated the potential interest of proton nuclear magnetic resonance spectroscopy in the evaluation of lungs in the early phase of reperfusion. Proton nuclear magnetic resonance spectroscopy gives information on mitochondria status and the level of reperfusion damage. This method could be interesting to assess grafts during the early period of reperfusion.
JAYLE ET AL NMRS FOR LUNG PRESERVATION EVALUATION
901
This work was supported by grants from the Fondation Pour la Recherche Me´ dicale, Paris, France. We thank Catherine Henry, Raymond Poncharaud, Nathalie Quellard, and Beatrice Fernandez for their technical assistance. We thank the Society Terumo for the donation of Oxygenator Capiox SX10 (Terumo Corp, Tokyo, Japan).
References 1. Cooper JD. Lung transplantation—a new era. Ann Thorac Surg 1987;44:447–8. 2. Christie JD, Bavaria JE, Palesky HI, et al. Primary graft failure following lung transplantation. Chest 1998;114:51–60. 3. Schulman LL. Perioperative mortality and primary graft failure. Chest 1998;114:7–8. 4. Hosenpud JD, Novick RJ, Bennet LE, et al. The registry of the International Society for Heart and Lung transplantation: thirteenth official report, 1996. J Heart Lung Transpl 1996;15: 655. 5. Breda MA, Hall TS, Stuart S, et al. Twenty-four hour lung preservation by hypothermia, and leukocyte depletion. J Heart Transplant 1985;4:325–9. 6. Novick RJ, Gehman KE, Ali IS, Lee J. Lung preservation: the importance of endothelium and alveolar type II cell integrity. Ann Thorac Surg 1996;62:302–14. 7. Lefer AM, Lefer DJ. Pharmacology of the endothelium in ischemia-reperfusion and circulatory shock. Ann Rev Pharmacol Toxicol 1993;33:71–90. 8. Sakamaki F, Hoffmann H, Muller C, Dienemann H, Messmer K, Schildberg FW. Reduced lipid peroxidation and ischemia reperfusion injury after lung transplantation using low potassium dextran solution for lung preservation. Am J Respir Crit Care Med 1997;156:1073–81. 9. Novick RJ, Possmayer F, Veldhuizen RAW, et al. Surfactant analysis and replacement therapy: a future tool of the lung transplant surgeon? Ann Thorac Surg 1991;52:1194 –200. 10. Neild GH, Foxall PJD, Lindon JC, Holmes EC, Nicholson JK. Uroscopy in the 21st century: high field NMR spectroscopy. Nephrol Dial Transpl 1997;12:404 –17. 11. Gibelin H, Eugene M, Hebrard W, Henry C, Carretier M, Hauet T. A new approach to the evaluation of liver graft function by nuclear magnetic resonance spectroscopy. A comparative study between Euro-Collins and University of Wisconsin solutions. Clin Chem Lab Med 2000;38:1133–6. 12. Hauet T, Goujon JM, Tallineau C, Carretier M, Euge`ne M. Early evaluation of renal reperfusion injury after prolonged cold storage using proton nuclear magnetic resonance spectroscopy. Br J Surg 1999;86:401–9. 13. Hopkinson DN, Bhabra MS, Hooper TL. Pulmonary graft preservation: a worldwide survey of current critical practice. J Heart Lung Transpl 1998;17:525–31. 14. Xiong L, Legagneux J, Wassef M, et al. Protective effects of Celsior in lung transplantation. J Heart Lung Transpl 1999; 18:320 –7. 15. Moriyasu K, McKeown PP, Novitzky D, Snow TR. Preservation of competent rabbit lung function after 30 hours of storage with low-potassium dextran solution. J Heart Lung Transpl 1995;14:75–9. 16. Wicomb WN, Hill JD, Avery J, Collins GM. Optimal cardioplegia and 24-hour heart storage with simplified UW containing polyethylene glycol. Transplantation 1990;49:261. 17. Zheng T, Lanza RP, Soon-Shiong P. Prolonged pancreas preservation using a simplified solution containing polyethylene glycol. Transplantation 1991;51:63–6. 18. Itasaka H, Burns W, Wicomb WN, Egaya H, Collins G, Esquivel CO. Modification of rejection by polyethylene glycol in small bowel transplantation. Transplantation 1994;57: 645–8. 19. Hauet T, Baumert H, Amor I, et al. Protection of autotransplantated pig kidneys from ischemia-reperfusion injury by polyethylene glycol. Transplantation 2000;70:1569 –75.
GENERAL THORACIC
Ann Thorac Surg 2003;76:896 –902
902
JAYLE ET AL NMRS FOR LUNG PRESERVATION EVALUATION
GENERAL THORACIC
20. Tokunaga Y, Wicomb WN, Garcia-Kennedy R, Esquivel CO, Collins GM. The immunosuppressive effect of polyethylene glycol in a flush solution for rat liver transplantation. Transplantation 1992;54:756 –8. 21. Ferrera R, Michel P, Hadour G, Rodriguez C, Ovize M, Phan Tan Luu R. An optimal experimental design for the development of preservative heart solutions. J Heart Lung Transpl 2002;21:260 –70. 22. Hauet T, Baumert H, Faure JP, et al. Beneficial effects of low-potassium and polyethylene glycol solution on renal lipid peroxidation during 48-hour cold storage and normothermic reperfusion. Transplant Proc 1998;30:2798 –9. 23. Collins GM, Wicomb WN, Levin BS, Verma S, Avery J, Hill JD. Heart preservation solution containing polyethylene glycol: an immunosuppressive effect? Lancet 1991;338:890 –1. 24. Sthulmetier KM, Lin Y. Camouflaging endothelial cells: does it prolong graft survival? Bioch Bioph Acta 1999;1428:177–90. 25. Qayami AK, Nikbakht-Sangari M, Godin DV, et al. The relationship of ischemia-reperfusion injury of transplanted
Ann Thorac Surg 2003;76:896 –902
26. 27.
28.
29. 30.
lung and the up-regulation of major histocompatibility complex II on host peripheral. J Thorac Cardiovasc Surg 1998; 115:978 –89. Wheeldon D, Sharples L, Wallwork J, English T. Donor heart preservation survey. J Heart Lung Transpl 1992;11:986 –93. Eugene M, Hauet T, Mothes D, et al. Beneficial effects of a low-potassium and polyethylene glycol solution on renal function and structure during 48-hour cold storage preservation. Transplant Proc 1997;29:2360 –2. Hauet T, Mothes D, Goujon JM, Carretier M, Eugene M. Protective effect of polyethylene glycol against prolonged cold ischemia and reperfusion injury: study in the isolated perfused rat kidney. J Pharmacol Exp Ther 2001;297:946 –52. Park KD, Kim YS, Han DK, et al. Bacterial adhesion on PEG modified polyurethane surfaces. Biomaterials 1998;19:851–9. Hauet T, Mothes D, Bon D, et al. Proton NMR spectroscopy as a novel approach to the monitoring of citrate, and trimethylamine-N-oxyde excretion after kidney preservation. Transplant Proc 1997;29:2323–5.