| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1995;60:1360-1366
© 1995 The Society of Thoracic Surgeons
Departments of Cardiothoracic Surgery, Anesthesiology, Medicine (Pulmonary and Critical Care), Pathology, and Surgery, University of Washington, Seattle, Washington
Accepted for publication May 23, 1995.
 |
Abstract |
|---|
 TopFootnotes Abstract IntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Methods. Eighteen rabbits were assigned to one of three groups. Groups 1 and 2 were subjected to one hour of in situ right hilar occlusion followed by 2 hours of reperfusion. Group 3 was subjected to identical surgical dissection but the right hilum was never occluded. Group 1 rabbits received saline solution (1 mL/kg) before hilar occlusion and group 2 rabbits, monoclonal antibody 60.3, a blocking antibody for the CD18 adhesion molecule on the neutrophil surface (2 mg/kg). In 3 of the antibody-treated rabbits, flow cytometry was performed on blood neutrophils before and after administration of the antibody and 120 minutes after reperfusion.
Results. The rabbits in groups 1 and 2 had significantly increased alveolar neutrophil infiltrate and increased pulmonary vascular resistance compared with the rabbits in group 3. However, there was no significant difference between group 1 (saline solution treated) and group 2 (antibody treated). Antibody treatment did not block migration of neutrophils into the alveoli. Flow cytometry of circulating neutrophils demonstrated that CD18 was upregulated after reperfusion and that CD18 was fully blocked after antibody treatment for the duration of the study.
Conclusions. We conclude that a 1-hour period of warm ischemia followed by reperfusion results in upregulation of CD18 but that emigration of the neutrophils into the alveoli is not CD18 dependent in this injury.
|
Introduction |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
The CD11/CD18 glycoprotein adhesion complex on leukocytes mediates ischemia and reperfusion–induced PMN firm adhesion and transendothelial migration as well as PMN–mediated vascular injury in a variety of tissues [11–14]. A monoclonal antibody (MAb 60.3) that binds to a functional epitope on CD18 blocks tissue injury after ischemia and reperfusion of the rabbit ear [13] and prevents most of the adverse effects associated with resuscitation from hemorrhagic shock in rabbits and rhesus monkeys [10, 11].
In a model of rabbit lung ischemia-reperfusion (24 hours of left pulmonary artery occlusion followed by reperfusion), pretreatment with MAb 60.3 improved flow after reperfusion and decreased the PMN influx into the alveoli [15]. However, lung ischemia was incomplete in this model; the bronchial circulation was not interrupted, ventilation was intact, and the pulmonary veins were open, permitting reflux venous flow [16]. In addition, the no-reflow phenomenon limited initial reperfusion to 6% to 8% of cardiac output (10). Thus, this model did not reflect the situation after global lung ischemia that occurs during clinical lung transplantation.
We tested the hypothesis that the lung injury after warm global ischemia and reperfusion requires the CD11/CD18 glycoprotein complex. We used the 
-CD18 MAb 60.3 to block PMN aggregation and adherence to endothelium in an effort to reduce subsequent lung damage.
|
Material and Methods |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Anesthesia and Ventilation
Cervical tracheostomy was performed, and an endotracheal tube with an inner diameter of 4.0 mm was placed. The lungs were ventilated with 100% oxygen with a tidal volume of 10 mL/kg at a respiratory rate of 30 breaths/min using a small-animal ventilator. Minute ventilation was adjusted according to arterial blood gas measurements to maintain arterial carbon dioxide tension between 30 and 50 mm Hg. Body temperature was maintained at 36° to 38°C with a heating blanket and a neonatal artificial nose humidifier (Vital Sign HCH), placed in the ventilator circuit just proximal to the endotracheal tube.
Instrumentation
Catheters were placed in the superior vena cava and femoral artery. The arterial catheter had a thermistor at its tip for thermodilution CO determinations. A median sternotomy was performed, and the right and left hili were isolated by dividing the inferior pulmonary ligaments and associated connective tissue. The mediastinal lobe was included with the right lung during this study. Vessi-loop Rumel tourniquets were placed loosely around both hili for later hilar occlusion. The pericardium was opened, and a pulmonary artery catheter was placed using a needle and guidewire through the right ventricular outflow tract across the pulmonary valve.
Study Groups
Eighteen rabbits were randomly assigned to three groups (n = 6 per group). Groups 1 and 2 were subjected to 1 hour of right hilar occlusion (hilar tourniquet applied at end-expiration) followed by 2 hours of right lung reperfusion. The left hilum was occluded at the same time the right lung was reperfused. Group 1 received saline solution (1 mL/kg) before the hilar occlusion and group 2, MAb 60.3 (2 mg/kg) in a total volume of 1 mL/kg. Investigators were blinded as to whether the rabbits received the MAb or saline solution treatment. After 2 hours of reperfusion, the rabbits were killed with an overdose of pentobarbital. Group 3 was subjected to identical surgical dissection, but the right hilum was not occluded and the rabbits did not receive antibody. After 1 hour of two-lung ventilation, the left hilum was occluded, just as in groups 1 and 2, followed by 2 hours of right-lung ventilation prior to death.
Blood Gas and Hemodynamic Assessments
Arterial blood gases, thermodilution CO, central venous pressure, and pulmonary artery pressure were measured at baseline and 15, 30, 60, 90, and 120 minutes after reperfusion. Compliance was calculated as the slope of the linear regression line of the plot of volume (10, 20, 30, and 40 mL) versus pressure (measured). Total pulmonary resistance (dynes•s•cm-5) was calculated as 80 x mean pulmonary artery pressure (mm Hg)/CO (L/min). After the death of the animal, the right middle lobe was removed for histologic study and lung water measurements, and the remainder of the right lung was lavaged.
Lung Water
Lung wet to dry weight ratios were calculated after the middle lobe was dried until weight was constant in an oven at 80°C. The weight was corrected for erythrocyte mass as will be described. Sixty minutes prior to termination, a 3.5-mL sample of the rabbit's blood was collected in 0.5 mL of acid-citrate-dextrose solution. Ten microcuries of chromium 51 (New England Nuclear No. NEZ-0301) was added, and the sample was agitated for 30 minutes and then centrifuged at 3,000 rpm for 15 minutes. The cells were then resuspended with 3 mL of normal saline solution, centrifuged, and resuspended with 3 mL of normal saline solution. Five minutes prior to termination, the labeled erythrocytes were injected into the rabbit. After weighing, the lung and blood samples were counted in a gamma counter. The fraction of the lung weight attributable to intravascular blood volume was calculated, and the wet to dry ratio was corrected to a blood-free value.
Lung Lavage
The right upper lobes and the right lower lobes were lavaged with 15 mL of normal saline solution instilled and withdrawn three times. Total lavage leukocyte counts were determined using a hemocytometer, and differential counts were performed on Wright-stained cells prepared by cytospinning. Protein content was determined using the Pierce BCA protein assay.
Pathology
Specimens from the right middle lobe were placed in 2% paraformaldehyde and 4% glutaraldehyde in 0.1 mol/L cacodylate buffer for at least 24 hours. The tissues were then dehydrated and embedded in paraffin. Five-micrometer sections were mounted on glass slides, deparaffinized, and stained with hematoxylin and eosin. The pathologist was unaware of group treatment. A minimum of ten fields were randomly examined by light microscopy at x400, and the lungs were ranked on the basis of severity of edema and alveolar infiltrate as previously described by us [17].
Flow Cytometry
In 3 of the MAb 60.3–treated rabbits, indirect one-color flow cytometry with fluorescein isothiocyanate labeling was used to analyze PMN CD18 expression and saturation with antibody. Whole-blood samples were collected in 3.8% sodium citrate at baseline, after MAb 60.3 treatment (before hilar occlusion), and 120 minutes after reperfusion. Lavage samples were collected in normal saline solution at 120 minutes after reperfusion. Samples were immediately cooled to 4°C and subsequently prepared at that temperature.
Erythrocytes were lysed in NH4Cl/KHCO3/tetrasodium EDTA (ethylenediaminetetraacetic acid) solution, and leukocytes were recovered by centrifugation. The leukocytes were washed in phosphate-buffered saline solution (divalent cation-negative) followed by trypan blue exclusion to confirm viability greater than 95%. After incubation with heat-inactivated bovine serum, leukocytes were incubated with either MAb 60.3 or murine IgG (negative control). Cells were washed and then incubated with secondary antibody (fluorescein isothiocyanate–conjugated goat anti-mouse IgG). Cells incubated with secondary antibody in the absence of primary antibody were used as a second negative control. After a final wash, leukocytes were fixed in 1% paraformaldehyde and stored at 4°C in the dark.
Flow cytometric analysis was performed on an EPICS 750 flow cytometer (Coulter) equipped with an MDADS data-acquisition system (Coulter). Polymorphonuclear leukocytes were identified by appropriate gating of cell size and granularity on the basis of characteristic forward and perpendicular light scatter observations. Histograms of cell number versus fluorescence intensity on a three-decade log scale were generated using at least 10,000 cells per sample.
Statistical Analysis
Statistical analysis was done using SPSS/PC+ one-way analysis of variance, unless otherwise specified, to assess between-group differences with least-significant difference testing used to identify specific differences. Paired t test was used for within-group data as specified below. Multivariate analysis of variance for repeated measures was used to assess differences over time. Log transformation was done for nonnormally distributed data if transformation increased the normality of the distribution. Data are presented as the mean with 95% confidence intervals. Log-transformed data are presented as the mean of the antilog with 95% confidence limits in brackets.
|
Results |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
). However, there were no significant differences between the saline solution–treated group (group 1) and the group treated with MAb 60.3 (group 2) for all end points measured. Two rabbits in the MAb 60.3–treated group became progressively hypoxemic and died 5 and 10 minutes before planned death.
|
|
). Protein concentration in the lavage fluid tended to differ between groups when groups 1 and 2 were combined and compared with group 3 (p < 0.06) (see Table 1).
Blood Gas Measurements
Arterial oxygenation was lower in groups 1 and 2 compared with group 3 (p < 0.05 by multivariate analysis of variance) (see Table 2). A progressive metabolic acidosis developed in groups 1 and 2 (see Table 2). It was partially corrected by increasing minute ventilation, but complete correction could not be done because maximal ventilation was being provided.
Hemodynamic Data
Cardiac output decreased over time in groups 1 and 2 but remained relatively stable in group 3 until 120 minutes (p < 0.05 by multivariate analysis of variance) (Table 3). Observation showed that the right ventricle became dilated in groups 1 and 2 compared with group 3. Total pulmonary resistance was significantly elevated after reperfusion in groups 1 and 2 compared with group 3 (p < 0.05 by multivariate analysis of variance) (see Table 3).
|
Lung Water and Pathology
Wet to dry ratios were 7.68 (6.05 to 9.73) and 8.06 (6.47 to 10.0) in groups 1 and 2, respectively, and 6.63 (5.15 to 8.54) in group 3 (p = 0.14). All three were higher than our laboratory control of 5.35 ± 0.12 for unmanipulated normal rabbit lung. Histology demonstrated areas of moderate to severe alveolar edema and leukocyte infiltration. However, there were no identifiable differences between the three groups in regard to extent of leukocyte infiltrate or degree of alveolar flooding.
Flow Cytometry
Flow cytometry of circulating PMNs collected from three MAb 60.3–treated animals demonstrated complete saturation of surface CD18 with MAb 60.3 both on samples taken after antibody treatment (before hilar occlusion) and at 120 minutes after reperfusion (Fig 1A–1D). Lavage PMNs collected from the same animals at 120 minutes after reperfusion were also shown to be completely saturated with MAb 60.3 (Fig 1E, 1F). Circulating PMNs demonstrated upregulation of CD18 at 120 minutes as seen by the increased fluorescence intensity (rightward shift) (see Fig 1C versus 1B).
|
|
Comment |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Polymorphonuclear leukocytes have been implicated in tissue injury after ischemia and reperfusion in a variety of different organs [13, 18]. In this study, we examined the role of the CD11/CD18 complex in lung ischemia-reperfusion. Monoclonal antibody 60.3 is directed to a functional epitope on CD18 and blocks PMN adhesion to its endothelial ligand, intercellular adhesion molecule-1 (ICAM-1) [14, 19]. Treatment with MAb 60.3 inhibits PMN migration in peritonitis, attenuates peripheral ischemia-reperfusion injury, and reduces organ injury after hemorrhagic shock and resuscitation [11–13, 20]. Pretreatment with MAb 60.3 reduced lung injury, endotoxin-induced PMN emigration, and gram-negative sepsis–induced alveolar capillary membrane injury [20–22].
Our group [15] previously demonstrated in rabbits that 24 hours of left pulmonary artery occlusion followed by reperfusion led to increased pulmonary vascular resistance and increased PMN emigration into the alveoli. This response was partially blocked by MAb 60.3. In contrast, the current study reports a CD18-independent pathway of pulmonary PMN emigration after lung ischemia-reperfusion. One explanation for this discrepancy is a differing time course of injury that may affect endothelial expression of ICAM-1, the principal adhesion ligand for PMN CD18. Expression of ICAM-1 is transcriptionally regulated and after inflammatory stimulation of vascular endothelium in vitro, reaches its maximal expression at 6 to 8 hours [23]. Our previous report [15] documenting CD18-dependent pulmonary PMN emigration used a 24-hour reperfusion period, providing ample time for ICAM-1 upregulation. The current protocol used only a 2-hour reperfusion period, possibly limiting ICAM-1 upregulation, although it has been previously reported that rabbit lung ischemia followed by only 2-hour reperfusion significantly increased endothelial ICAM-1 expression and resulted in lung injury that is both PMN mediated and ICAM-1 dependent [24].
Although we did not attempt to document lung ICAM-1 expression in the current experiments (eg, by immunocytochemistry), other investigators have shown that in contrast to the prolonged time course of ICAM-1 expression after inflammatory stimulation in vitro, the time course of ICAM-1 expression or function or both in vivo is more variable and more rapid than in in vitro experiments. In the rat, anti–ICAM-1 MAb significantly reduced lung injury and neutrophil emigration 30 minutes after intravenous administration of cobra venom factor [25], 1 hour after intratracheal administration of tumor necrosis factor [26], 4 hours after IgA immune complex injury [27], and 4 hours after hind limb ischemia-reperfusion [28]. We cannot confirm that lung ICAM-1 expression was increased in our current experiments, but on the basis of these previous reports, it is quite possible that adequate ICAM-1 to promote CD18-dependent adherence was present by 2 hours after reperfusion. Nevertheless, we found PMN emigration under these conditions to be CD18 independent.
A second explanation for the discrepancy in CD18 requirements in this and previous reports [15, 24] of lung ischemia-reperfusion is a difference in the method of producing ischemia. In the CD18-dependent injury model [15, 24], pulmonary ischemia was achieved by intravascular pulmonary artery occlusion, which left the bronchial (systemic) circulation to the lung intact. The current study used hilar occlusion, thereby completely interrupting pulmonary and systemic flow to the involved lung. These observations suggest that the bronchial circulation may play an important role in CD18-dependent PMN emigration after ischemia-reperfusion.
Our findings of CD18-independent (current study) and CD18-dependent [15] PMN emigration add to previous reports that PMN emigration in the lung is also stimulus dependent, time dependent, or both. CD18-independent mechanisms play a role in the lung in certain bacterial infections [29], intratracheal C5a administration [30], and early after endotoxin administration [31], whereas CD18-dependent mechanisms play a role late (24 hours) after endotoxin administration [29, 31]. In addition, studies of lung ischemia and reperfusion using pulmonary artery occlusion suggest the importance of CD18-dependent mechanisms of PMN migration [15, 24, 32]. In contrast, whole-body ischemia-reperfusion injury (somewhat analogous to the pulmonary hilar occlusion method of pulmonary ischemia used in the current study) appears to induce CD18-independent PMN migration in the lung [12].
Confirmation that our technique of hilar occlusion resulted in complete lung ischemia was accomplished by radiolabeled microsphere injections in 3 additional rabbits. Reperfusion was with total CO to simulate the transplantation situation in which 70% to 80% of CO goes to the donor side because of preexisting pulmonary hypertension [33]. This resulted in systemic hemodynamic changes after reperfusion with decreased CO, increased total pulmonary resistance, and a dilated right heart. Further, we evaluated warm ischemia only, when, in, fact donor lungs undergo a short period of warm ischemia followed by a longer period of cold ischemia. We cannot rule out the possibility that reperfusion after cold ischemia may be partially or completely CD18 dependent.
Development of mild edema even in group 3 rabbits was unexpected and may have resulted from a combination of handling the lung and possible lymphatic disruption. In addition, flow through this lung was significantly increased by left hilar occlusion, resulting in recruitment of pulmonary vasculature in the right lung and providing a greater lung vascular surface area and thus an increased permeability–surface area product. This increases microvascular fluid filtration in the lung and could increase the wet to dry ratio.
Given the marked differences in lavage data leukocyte differential counts, pulmonary resistance, and the hemodynamic effects of reperfusion, we were surprised by the lack of obvious histologic differences. It is conceivable that the middle lobes examined microscopically were less affected than the lower lobes, which receive most of the flow. The lack of alveolar damage may also be explained by the relatively brief period of reperfusion. We did not extend the reperfusion period longer because pilot studies resulted in a high mortality when reperfusion was extended beyond 2 hours. The short observation period may have resulted in missed late PMN infiltration. Also, subjective grading of alveolar infiltrate and edema is presumably less sensitive than quantitative techniques.
Flow cytometry demonstrated that surface expression of CD11/CD18 on circulating PMNs was increased during the time course of reperfusion. This suggests that PMNs were activated during ischemia or reperfusion or both. The dose of 2 mg/kg of MAb 60.3, as expected, saturated all CD18 receptors, as determined by the absence of increased fluorescence with further addition in vitro of MAb 60.3 to PMNs of treated rabbits. The PMNs present in the alveolar lavage fluid were also saturated with MAb 60.3, which indicated that despite CD18 upregulation, PMN emigration was independent of CD18. Both E selectin [34] and P-selectin [35] have been implicated in PMN adhesion pathways in the injured lung and may serve as alternative pathways of CD18-independent PMN adhesion in the lung. Because selectins are not known to be involved in endothelial transmigration of PMNs, however, the alternative pathway of PMN emigration in the current studies is unclear.
Increased total pulmonary resistance after reperfusion is a frequently observed phenomenon. Monoclonal antibody 60.3 treatment limited the increase in pulmonary resistance when the pulmonary artery was occluded for 24 hours [15]. However, the right heart dilatation and decreased CO do not seem completely explainable on the basis of increased pulmonary resistance. Another possible mechanism is that tumor necrosis factor released after pulmonary artery occlusion and reperfusion [36] caused myocardial depression as demonstrated by Heard and colleagues [37]. We did not measure tumor necrosis factor, and this hypothesis remains speculative.
In summary, warm global lung ischemia and reperfusion led to acute PMN emigration into the rabbit lung with increased pulmonary resistance and decreased CO. Ischemia-reperfusion also caused upregulation of the adhesion molecule CD18 (part of the CD11/CD18 complex) on the PMN surface. However, blocking CD18-dependent adherence with MAb 60.3 did not prevent any of the observed changes of cardiopulmonary injury. Furthermore, PMNs in the alveoli were saturated with MAb 60.3 and must, therefore, migrate by a CD18-independent mechanism under these conditions of lung ischemia and reperfusion.
|
Acknowledgments |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
Footnotes |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
regulates in vivo intrapulmonary expression of ICAM-1. Am J Pathol 1993;142: 1739–49.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  Y.-T. Lu, P.-G. Chen, and S. F. Liu Time course of lung ischemia-reperfusion-induced ICAM-1 expression and its role in ischemia-reperfusion lung injury J Appl Physiol, August 1, 2002; 93(2): 620 - 628. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. C. Ridger, B. E. Wagner, W. A. H. Wallace, and P. G. Hellewell Differential Effects of CD18, CD29, and CD49 Integrin Subunit Inhibition on Neutrophil Migration in Pulmonary Inflammation J. Immunol., March 1, 2001; 166(5): 3484 - 3490. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Chello, P. Mastroroberto, E. Malta, F. Cirillo, and V. Celi Inhibition by dipyridamole of neutrophil adhesion to vascular endothelium during coronary bypass surgery Ann. Thorac. Surg., May 1, 1999; 67(5): 1277 - 1282. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Halldorsson, M. Kronon, B. S. Allen, K. S. Bolling, T. Wang, S. Rahman, H. Feinberg, and R. S. Hartz Controlled Reperfusion After Lung Ischemia: Implications For Improved Function After Lung Transplantation J. Thorac. Cardiovasc. Surg., February 1, 1998; 115(2): 415 - 425. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. V. Parsey, R. M. Tuder, and E. Abraham Neutrophils Are Major Contributors to Intraparenchymal Lung IL-1{beta} Expression After Hemorrhage and Endotoxemia J. Immunol., January 15, 1998; 160(2): 1007 - 1013. [Abstract] [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 |
