| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1998;65:187-192
© 1998 The Society of Thoracic Surgeons
Department of Cardiothoracic Surgery, Wythenshawe Hospital, Manchester, United Kingdom
Accepted for publication July 3, 1997.
Dr Hooper, Department of Cardiothoracic Surgery, Wythenshawe Hospital, Southmoor Rd, Manchester M23 9LT, United Kingdom.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Methods. Rat lungs were reperfused ex vivo using a parabiotic animal after 0-hour (groups A through C) or 24-hour (groups D through G) storage. Reperfusion pressure was either physiologic (groups A through D) or reduced by 50% for a specified time (groups E through G). The duration of reperfusion was 5 minutes (groups A, D, and E), 10 minutes (groups B and F), or 30 minutes (groups C and G), at which time endothelial permeability was measured through iodine 125–labeled albumin leakage and neutrophil sequestration through tissue myeloperoxidase activity.
Results. Graft function in group D deteriorated rapidly, whereas groups E through G performed at control levels. Albumin leakage was significantly elevated in group D; with controlled reperfusion, it was elevated after 5 minutes (group E) but had returned to baseline at 10 minutes (group F) and 30 minutes (group G). Myeloperoxidase levels were not significantly different between groups.
Conclusions. Endothelial permeability is transiently elevated in the early phase of lung graft reperfusion. Initial low-pressure reperfusion may be protective by preventing irreversible edema formation during this period.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
In previous studies, we [5][6] have established that rat lung graft function can be significantly improved by lowering initial reperfusion pressure. Whereas reperfusion using physiologic pressure resulted in poor function in grafts that had been stored at 4°C for 24 hours in University of Wisconsin (UW) solution or 6 hours in Euro-Collins solution, reduction of the reperfusion pressure by half for the first 10 minutes subsequently yielded function similar to nonischemic controls [6]. Moreover, when a shorter (5-minute) initial period of low-pressure reperfusion was used, function still deteriorated. It would appear, therefore, that controlled reperfusion may either favorably alter the progression of I-R injury or somehow protect lung grafts from the consequences of this injury during a critical early phase. We hypothesized that the mechanism may involve beneficial alteration of edema-forming forces through the lowering of intravascular hydrostatic pressure during a transient early phase of increased endothelial permeability.
In the current study, we set out to measure endothelial permeability during different stages of reperfusion (0 to 5, 5 to 10, and 25 to 30 minutes) in lung grafts stored for 0 or 24 hours and reperfused either at physiologic pressure or with a 50% reduction in pressure for an initial 10-minute period. Further, we measured myeloperoxidase activity in the grafts so as to assess changes in neutrophil sequestration under these conditions.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Male Sprague-Dawley rats (Charles River Laboratories, Kent, UK) weighing 350 to 420 g were used. Anesthesia was induced with halothane and maintained with intraperitoneal administration of sodium pentobarbital, 100 mg/kg. Harvard rodent ventilators (Harvard Apparatus, Kent, UK) were used for animal and graft ventilation.
Donor Procedure
Lung donors were ventilated through a tracheostomy with room air using 60 breaths/min, 10 mL/kg tidal volume, and 3 cm H2O positive end-expiratory pressure. After a median sternotomy, 500 units of heparin sodium was administered, a left atriotomy was performed, and a cannula was placed in the pulmonary artery (PA). The lungs were then flushed through this cannula with 60 mL/kg of 4°C UW solution (DuPont Pharmaceuticals, Letchworth Garden City, UK), delivered using 25-cm hydrostatic pressure. The heart-lung block was excised and submerged in UW solution for storage.
Reperfusion
After excision of the left lung and postcaval lobe, grafts were reperfused in a warmed chamber for 1 hour with deoxygenated blood from a support animal and ventilated with room air using 30 breaths/min, 10 mL/kg tidal volume, and 3 cm H2O positive end-expiratory pressure. The support animals were cannulated through a median sternotomy. A 16-gauge cannula was passed through the right superior vena cava and right atrium so that its tip lay in the inferior vena cava. Blood drawn from the inferior vena cava was delivered hydrostatically to the grafts, which were situated below the support animals, the vertical distance between the animal and the graft normally being such that the reperfusion pressure generated approximated the physiologic PA pressure of this species (18 to 20 mm Hg). Effluent draining from the left atrium of the graft was collected and returned by a pump (variable-speed peristaltic pump; Harvard Apparatus) to the right atrium of the support animal through an 18-gauge cannula placed through the left superior vena cava. The support animals therefore functioned as physiologic deoxygenators. The circuit tubing and reperfusion chamber were water-jacketed to maintain blood temperature at 38°C, and support animals were placed on a warming blanket (Harvard Apparatus). Blood obtained from additional animals (1 donor per one or two experiments) and 0.9% saline solution were used to prime the circuit and to replace losses.
During reperfusion, the stability of the support animals was assessed by monitoring core temperature, heart rate, and degree of filling of the heart and great vessels and by acid-base and gas tension analysis of venous blood sampled from the circuit. The latter also allowed monitoring of graft reperfusate oxygen tension (PO2).
Experimental Protocol
Seven groups of experiments (groups A through G, n = 5 each) were carried out. Groups A, B, and C were nonischemic controls in which immediately after flush and explantation, grafts were reperfused for 5, 10, or 30 minutes, respectively, using physiologic reperfusion pressure. In groups D through G, grafts were stored at 4°C in UW solution for 24 hours. Group D grafts were reperfused for 5 minutes using physiologic pressure. In groups E, F, and G, controlled reperfusion was used. Grafts in groups E and F were reperfused for 5 and 10 minutes, respectively, at 50% pressure, achieved by halving the vertical distance between the grafts and support animals. In group G, reperfusion was at 50% pressure for the first 10 minutes and at full physiologic pressure for a further 20 minutes.
The study design is summarized in Table 1. Physiologic-pressure reperfusion alone of stored grafts was not extended beyond 5 minutes because by this time there was invariably macroscopic evidence of considerable edema formation. As a result, subsequent measurement of endothelial permeability using albumin leakage would have been difficult to interpret.
|
Permeability
Albumin leakage was used as an index of endothelial permeability. One milliliter of iodine 125–labeled human serum albumin (Amersham Intl, Little Chalfont, UK) containing approximately 1 µCi was added to the circuit 5 minutes before the termination of reperfusion, ie, at the beginning of reperfusion in groups A, D, and E, at 5 minutes in groups B and F, and at 25 minutes in groups C and G. At the end of the reperfusion period, a 1-mL sample of blood was taken from the graft effluent and weighed. All extrapulmonary tissue was dissected from the graft, and the lung was blotted dry gently and weighed. The graft and blood sample were counted separately for 1 minute each in a gamma counter. The albumin leak index was then calculated as the ratio counts · min-1 · g-1 of lung/counts · min-1 · g-1 of blood to give a measure of extravasation of albumin during the relevant 5-minute period.
Myeloperoxidase Assay
Myeloperoxidase occurs ponderantly in neutrophils, and it is widely used as a quantitative marker of neutrophil sequestration in lung tissue [7][8]. The assay was adapted from a previously described method [9]. After completion of reperfusion and counting, the lung grafts were snap-frozen in liquid nitrogen until the time of assay. Samples weighing 120 mg were homogenized in 1 mL of citrate-phosphate buffer with pH 5 (BDH/Merck, Poole, UK) at 4°C. The homogenate was sonicated to disrupt neutrophil granules and solubilize myeloperoxidase. The resulting suspension was centrifuged at 13,000 rpm for 6 minutes. One hundred–microliter samples of the supernatant diluted with citrate-phosphate buffer were placed in a multiwell plate, as were 100-µL aliquots of purified bovine myeloperoxidase standard (Sigma, Poole, UK), also diluted with buffer. To each well, 100 µL of a solution prepared using 1 mL of 2,2'-azino-bis(3-ethylbenzthiazoline-6-sulphonic acid) diammonium salt (ABTS) (Sigma), 1 µL of 30% hydrogen peroxide (BDH/Merck), and 50 mL of citrate-phosphate buffer was added. ABTS is a substrate for myeloperoxidase [10]. The plates were read at 37°C and 405 nm when the top dilution of the myeloperoxidase standard gave a change in absorbance of 1. Myeloperoxidase activity in the samples was calculated from their relative change in absorbance and the known activity of the enzyme standard.
Statistical Analysis
Data are expressed as the mean ± the standard error of the mean. Multiple comparisons were made by one-way analysis of variance followed by the Bonferroni post hoc test if differences were found. Comparison of two means was made by paired or unpaired Student t test as appropriate, and two-tailed p values were used. Values of p that were less than 0.05 were considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Graft Function
Function within the nonischemic control groups was stable and similar at corresponding time points. Hence, there were no significant differences in graft blood flow (Fig 1),
mean PA pressure (Fig 2),
PO2 (oxygenation) (Fig 3),
or peak airway pressure (Fig 4)
between groups A, B, and C at 5 minutes or between groups B and C at 10 minutes.
|
|
|
|
Within the controlled-reperfusion groups, function was similar at corresponding times, with no significant differences in blood flow (see Fig 1), mean PA pressure (see Fig 2), PO2 (see Fig 3), or peak airway pressure (see Fig 4) between groups E, F, and G at 5 minutes or between groups F and G at 10 minutes. During the first 10 minutes of reperfusion in groups F and G, despite the hydrostatic perfusion pressure being constant, there was a steady increase in blood flow (2.5 minutes versus 10 minutes: p < 0.01 for group F, p < 0.05 for group G) without any increase in PA pressure. The measured PA pressure, determined by a combination of hydrostatic reperfusion pressure, pulmonary vascular resistance, and positive-pressure ventilation, was similar to that in the control groups. With the subsequent increase in reperfusion pressure to physiologic levels, function remained similar to controls: at 30 minutes, blood flow in group G was 11.2 ± 0.8 mL/min versus 12.5 ± 0.6 mL/min in group C (see Fig 1), mean PA pressure was 15.1 ± 0.7 mm Hg versus 13.1 ± 0.2 mm Hg, respectively (see Fig 2), PO2 was 142 ± 4 mm Hg versus 137 ± 3 mm Hg, respectively (see Fig 3), and peak airway pressure was 14.3 ± 0.7 mm Hg versus 12.3 ± 0.9 mm Hg, respectively (see Fig 4) (p = not significant for all).
Permeability
The albumin leak index in nonischemic grafts was similar at 5, 10, and 30 minutes (groups A, B, and C, respectively; p = not significant) (Fig 5).
At 5 minutes, the albumin leak index in group A was 0.157 ± 0.015; in stored grafts, it was significantly higher with physiologic-pressure reperfusion in group D (0.930 ± 0.122; p < 0.001) and with low-pressure reperfusion in group E (0.491 ± 0.027; p < 0.01). At 10 minutes, the albumin leak index in low-pressure perfused stored grafts (group F) had come down to control (group B) levels (0.217 ± 0.005 versus 0.145 ± 0.017, respectively; p = not significant). With a subsequent 20 minutes of physiologic-pressure reperfusion, the albumin leak index remained similar to controls, being 0.174 ± 0.015 in group G and 0.157 ± 0.007 in group C (p = not significant).
|
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Physiologic-pressure reperfusion of stored grafts for 5 minutes (group D) resulted in significantly more albumin leakage than in controls. This was consistent with functional deterioration and macroscopic evidence of edema in the grafts. For this reason, albumin leakage at later time points was not measured in this group, as the results would not have been meaningful. With reperfusion pressure halved, there was a smaller but still significant increase in albumin leakage during the first 5 minutes (group E) compared with controls, despite the PA pressure being lower, but this was not associated with irreversible edema formation or functional impairment. At the same reperfusion pressure, albumin leakage fell to control levels between 5 and 10 minutes (group F). Even with the subsequent increase in reperfusion pressure to physiologic levels (group G), albumin leakage remained similar to controls at 30 minutes.
These data suggest that endothelial permeability in the stored grafts was increased during the first 5 minutes of reperfusion but had returned to normal by 10 minutes. Lowering reperfusion pressure for the first 10 minutes may protect lung grafts by reducing hydrostatic forces, thus preventing irreversible formation of edema. Recovery of endothelial permeability barrier function by 10 minutes may then allow tolerance of higher hydrostatic pressures. This is supported by our finding [6] previously that a shorter initial period of 5 minutes of low-pressure reperfusion does not prevent graft failure.
The steady increase in blood flow in groups F and G grafts during the first 10 minutes of reperfusion may also be a sign of recovery of endothelial function. Endothelial dysfunction plays a critical role in the pathogenesis of I-R injury and is marked by decreased production of protective autocoids such as nitric oxide, prostacyclin, and adenosine during reperfusion [11]. The observed hemodynamic improvement may be secondary to restoration of endothelial production of these agents, all of which are powerful vasodilators.
Although our data suggest that the mechanism for the beneficial effect of initial low-pressure reperfusion may be through reduction of hydrostatic pressure, the possibility that physiologic-pressure reperfusion may actually increase the severity of endothelial injury cannot be excluded. In a rat lung transplant model [12], it was found that after 48 or 72 hours of hypothermic storage, endothelial continuity was maintained with only some small intercellular gaps appearing after 72 hours, whereas after just 5 minutes of subsequent reperfusion, there was patchy denudation of endothelium and loss of connections between endothelial cells as well as perivascular and alveolar edema. Injury to endothelial cells during ischemia and early reperfusion may weaken their attachments and make them susceptible to greater shear stresses associated with relatively high pressure reperfusion, leading to loss of endothelial continuity and edema formation. Such disruptive effects of shear stresses may supersede their beneficial actions, which include stimulation of endothelial nitric oxide production and deterrence of interaction of circulating neutrophils with the endothelium [8].
Increased endothelial permeability is a well-recognized result of I-R injury, and aside from mechanical trauma, generation of oxidants during reperfusion is thought to be a causative factor [4]. Indeed, increased permeability is frequently used as a measure of I-R injury in lungs [7][8]. We chose to use radiolabeled albumin leakage for measurement of permeability in preference to the alternative technique of calculating capillary filtration coefficients from changes in graft weight in response to a sudden increase in venous pressure [4][8] because the latter technique requires an initial period of stable isogravimetric reperfusion, whereas we were interested in rapid, early changes in permeability.
The absence of any difference in myeloperoxidase activity between groups at 5 minutes is not surprising. The involvement of neutrophils in I-R injury requires the expression of adhesion molecules on the endothelium. This is triggered by endothelial damage incurred during ischemia and especially at reperfusion, when a burst of oxygen-derived free radicals is generated [13]. Of the adhesion molecules, some (eg, P-selectin) are stored in cytoplasmic bodies and can be rapidly translocated to the endothelium, but others (eg, E-selectin, ICAM-1) require synthesis de novo, and so their expression takes longer [14]. Eppinger and colleagues [7] showed that lung I-R injury (increased permeability) may be biphasic, with a peak at 30 minutes of reperfusion without any increase in myeloperoxidase activity and another peak at 4 hours, this time with evidence of increased neutrophil sequestration. The initial phase of injury was unaffected by neutrophil depletion. Other studies [15][16] have also shown that high-permeability lung I-R injury can occur in the absence of neutrophils. Ultrastructural studies [17][18] of transplanted lungs after 4 hours of reperfusion showed resolution of edema present at the end of the preservation period but a significantly increased presence of neutrophils. There is no doubt that neutrophils play an important part in I-R injury, but they may not be involved in the early changes in permeability observed in the current study. It should, however, be borne in mind that the use of a paracorporeal circuit can reduce the number of circulating neutrophils, and this may have influenced our results.
Low-pressure reperfusion may potentially be beneficial through a number of mechanisms, eg, by avoiding abrupt changes in pH, temperature, or oxygenation. Recent studies [19][20] have also shown that high-flow reperfusion produces more lung injury after preservation with high-potassium than low-potassium solutions; the UW solution used in this study had a high potassium content. However, the consequences of controlled reperfusion on hydrostatic pressure and shear stress are probably of greatest importance, and our studies suggest that this may be through the provision of a window of protection during which the endothelium recovers from a period of increased permeability. In clinical transplantation, a short initial period of low-pressure reperfusion would be relatively simple to apply and is unlikely to have any detrimental effects. In single-lung transplantation, the graft frequently receives the majority of the cardiac output, particularly when there is preexisting pulmonary hypertension, whereas in sequential bilateral transplantation, the entire cardiac output can be to the first lung graft during implantation of the second one. In these situations, controlled reperfusion may be of even more importance in minimizing early graft dysfunction.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
We thank Dr Paul Brenchley, Department of Immunology, and Dr Harbans Sharma and Ann-Marie Smith, Department of Medical Biophysics, University of Manchester, for their advice and assistance.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Casiraghi, J. R. Tatreau, J. B. Abano, J. W. Blackwell, L. Watson, K. Burridge, S. H. Randell, and T. M. Egan In vitro modeling of nonhypoxic cold ischemia-reperfusion simulating lung transplantation J. Thorac. Cardiovasc. Surg., September 1, 2009; 138(3): 760 - 767. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Lang, C. Aigner, G. Winkler, K. Shkirdladze, W. Wisser, G. Dekan, M. Tamura, G. Heinze, D. Van Raemdonck, and W. Klepetko Prolonged venoarterial extracorporeal membrane oxygenation after transplantation restores functional integrity of severely injured lung allografts and prevents the development of pulmonary graft failure in a pig model. J. Thorac. Cardiovasc. Surg., June 1, 2009; 137(6): 1493 - 1498. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Aigner, W. Wisser, S. Taghavi, G. Lang, P. Jaksch, D. Czyzewski, and W. Klepetko Institutional experience with extracorporeal membrane oxygenation in lung transplantation Eur. J. Cardiothorac. Surg., March 1, 2007; 31(3): 468 - 474. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. M. Marta, C. Aigner, and W. Klepetko Split lung transplantation with intraoperative extracorporeal membrane oxygenation (ECMO) support MMCTS, August 9, 2005; 2005(0809): 984. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. L. Lau, G. A. Patterson, and S. M. Palmer Critical Care Aspects of Lung Transplantation J Intensive Care Med, March 1, 2004; 19(2): 83 - 104. [Abstract] [PDF]
|
|
|
|
 |
|
 C.L. Lau and G.A. Patterson Current status of lung transplantation Eur. Respir. J., November 16, 2003; 22(47_suppl): 57s - 64s. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. de Perrot, M. Liu, T. K. Waddell, and S. Keshavjee Ischemia-Reperfusion-induced Lung Injury Am. J. Respir. Crit. Care Med., February 15, 2003; 167(4): 490 - 511. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. R Rega, E. J Vandezande, N. C Jannis, G. M Verleden, T. E Lerut, and D. E. Van Raemdonck The role of leukocyte depletion in ex vivo evaluation of pulmonary grafts from (non-)heart-beating donors Perfusion, January 1, 2003; 18(1_suppl): 13 - 21. [Abstract] [PDF]
|
|
|
|
 |
|
 A. O. Halldorsson, M. T. Kronon, B. S. Allen, S. Rahman, and T. Wang Lowering reperfusion pressure reduces the injury after pulmonary ischemia Ann. Thorac. Surg., January 1, 2000; 69(1): 198 - 203. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. T.M. Tang, P. Geraghty, M. J. Dascombe, J. C. Jarvis, S. Salmons, and T. L. Hooper Nitroglycerine reduces neutrophil activation and acute damage in latissimus dorsi muscle grafts Ann. Thorac. Surg., December 1, 1998; 66(6): 2015 - 2021. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. J. Rooney and R. S. Bonser Controlled perfusion protects lung grafts during a transient early increase in permeability Ann. Thorac. Surg., December 1, 1998; 66(6): 2155 - 2155. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
