|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Department of Internal Medicine (Drs. Cugno and Agostoni), IRCCS Maggiore Hospital; Angelo Bianchi Bonomi Hemophilia and Thrombosis Center (Dr. Coppola), Division of Cardiac Surgery (Drs. Biglioli and Alamanni), IRCCS Foundation Monzino, University of Milan, Milan, Italy; and Hypertension Division (Dr. Nussberger), University Hospital, Lausanne, Switzerland.
Correspondence to: Angelo Agostoni, MD, Department of Internal Medicine, University of Milan, Via Pace, 15, 20122 Milan, Italy; e-mail: intermed{at}mailserver.unimi.it
| Abstract |
|---|
|
|
|---|
Design: Prospective, observational study.
Setting: University hospital, cardiac surgery unit.
Patients and methods: Intra-arterial BP was monitored and serial blood samples were obtained from 27 patients undergoing CPB for cardiac surgery. We measured plasma bradykinin and parameters of coagulation, fibrinolysis, complement, contact system, and the cytokine tumor necrosis factor (TNF).
Results: Mean arterial pressure fell progressively until the end of CPB (- 18 mm Hg, p = 0.001) but returned to baseline by the end of surgery. The venous bradykinin level, normal in basal conditions (median, 1.90 fmol/mL), was increased (p = 0.001) from 15 min after the beginning of CPB (5.71 fmol/mL) to the end of the operation (7.07 fmol/mL), with a peak at the end of CPB (9.81 fmol/mL; p = 0.0001); it was normal at recovery 24 h later (2.81 fmol/mL). Bradykinin plasma levels fell 60% across the lung when the pulmonary circulation was fully restored while the patients were still receiving CPB. Activated-factor XII, thrombin-antithrombin complexes, prothrombin fragment F1 + 2, plasmin-antiplasmin complexes, C3a, and TNF increased significantly after the beginning of the surgical procedure, rising further during CPB, and remained elevated until the end of surgery, but they all returned to normal within 24 h. Changes in plasma bradykinin levels were not correlated with any of the other variables.
Conclusions: During CPB, there is a progressive increase of plasma bradykinin that is at least partially due to reduced catabolism as a consequence of shunting the lungs. The increase in bradykinin may contribute to the fall in BP.
Key Words: angiotensin-converting enzyme bradykinin metabolism coagulation complement extracorporeal circulation fibrinolysis hypotension
| Introduction |
|---|
|
|
|---|
In the present study, we used a recently developed sensitive and specific method10 to measure plasma levels of bradykinin in patients with coronary artery disease undergoing CPB. In view of the close links among the kinin system, hemostasis, and inflammation,8 changes in bradykinin levels were compared with activation of the contact system, coagulation, fibrinolysis, complement system, and cytokines, and analyzed in relation to changes in BP. Moreover, because bradykinin is assumed to be mainly inactivated in the pulmonary vascular tree,11 we evaluated its clearance during the restoration of the circulation across the natural lungs when extracorporeal circulation (ECC) was still ongoing. This study completes the preliminary data presented at the 15th International Conference on Kinins.12
| Materials and Methods |
|---|
|
|
|---|
In 21 patients, blood samples were obtained from a central venous catheter, and during CPB from the venous line of the bypass system. The sampling times were as follows: before surgery (baseline), after heparin infusion before ECC, 15 min after the beginning of ECC, at the end of ECC, at the end of surgery, and on the day after surgery (recovery). We compared changes in bradykinin with several parameters of the contact system (activated-factor XII [FXIIa], cleavage of HK), the coagulation cascade (prothrombin fragment F1 + 2 and thrombin-antithrombin [TAT] complexes), the fibrinolytic system (plasmin-antiplasmin [PAP] complexes), the complement system (C3a), and the cytokine tumor necrosis factor (TNF). These patients were disconnected from ECC after a weaning time of 30 ± 20 min. The weaning time, which follows the rewarming period, is defined as the time between aortic declamping and the end of ECC. This period is needed for the accomplishment of proximal anastomoses with partial clamping of the ascending aorta and then for the gradual restoration of satisfactory cardiac output.
The remaining six patients were studied in order to evaluate the generation and clearance of bradykinin by the artificial circuit and by the natural lung. In these six patients, we took paired samples from entrance and exit of the artificial circuit during CPB, and from the right atrium and aorta still during CPB after rapidly restoring the filling pressures and pulmonary blood flow.
Plasma bradykinin levels were measured specifically by radioimmunoassay after liquid-phase extraction and subsequent high-performance liquid chromatography, as described in detail.10 The detection limit was 0.2 fmol/mL. Intra-assay and interassay coefficients of variation were 18% at the low endogenous concentrations.
FXIIa was measured in trisodium citrated plasma with a sandwich enzyme-linked immunosorbent assay (ELISA) [Shield Diagnostic Ltd; Dundee, UK], which uses a specific mouse monoclonal antibody for human FXIIa (2/215) as capture antibody and a sheep polyclonal anti-FXIIa antibody conjugated to alkaline phosphatase as second antibody.13 Intra-assay and interassay coefficients of variation were 5.1% and 8.2%, respectively.
Cleavage of HK was assessed in plasma obtained from blood drawn into polypropylene tubes containing acid-citrate-dextrose (100 mM trisodium citrate, 67 mM citric acid, and 2% dextrose, pH 4.5), benzamidine (100 mM), 400 µg/mL hexadimethrine bromide, 2 mg/mL soybean trypsin inhibitor, 263 µM leupeptin and 20 mM aminoethylbenzenesulfonylfluoride (nine parts of blood into one part of anticoagulant).14 Sodium dodecyl sulfate-polyacrylamide gel electrophoresis and immunoblotting analysis was carried out by a method based on that described by Berrettini et al.15 After electrophoretic transfer of proteins from gel onto a polyvinylidene difluoride membrane (Immobilon; Millipore Corporation; Milford, MA), HK was identified with goat polyclonal anti-light chain HK (Nordic; Tilburg; Netherlands) and visualized with a biotinylated rabbit antigoat antibody (Sigma Chemical; St. Louis, MO). The apparent molecular masses of proteins were estimated by comparison with the high-molecular-weight protein markers from Bio-Rad Laboratories (Richmond, CA). With this method, native HK appears as a band with molecular weight of 130,000, and cleaved HK is represented by two bands with molecular weights of 107,000 and 98,000. The density of the bands was evaluated by computerized image analysis (Image Master; Pharmacia; Uppsala, Sweden). The amount of cleaved HK (bands with molecular weights of 107,000 and 98,000) was expressed as a percentage of the total HK (sum of the three bands).16
Prothrombin fragment F1 + 2 was assessed in citrated plasma with a sandwich ELISA (Enzygnost F1 + 2; Behring Diagnostics GmbH; Marburg, Germany). The method uses a rabbit antibody to human prothrombin fragment F1 + 2 as capture antibody and a rabbit peroxidase conjugated antiprothrombin as second antibody. Intra-assay and interassay coefficients of variation were 8%.
TAT complexes were measured in citrated plasma with a sandwich ELISA (Enzygnost TAT micro; Behring Diagnostics GmbH). The method uses a rabbit antibody against human thrombin as capture antibody and a rabbit peroxidase-conjugated antihuman antithrombin III. Intra-assay and interassay coefficients of variation were 6% and 9%, respectively.
PAP complexes were measured in citrated plasma with a sandwich ELISA (Enzygnost PAP micro; Behring Diagnostics GmbH). The method uses a rabbit antibody against human PAP as capture antibody and a rabbit peroxidase conjugated antihuman plasminogen. Intra-assay and interassay coefficients of variation were 3% and 6%, respectively.
C3a levels were measured in disodium ethylenediaminotetracetic acid plasma by radioimmunoassay (Amersham, England), according to Wagner and Hugli.17 Intra-assay and interassay coefficients of variation were 10% and 12%, respectively.
TNF levels were measured in disodium ethylenediaminotetracetic acid
plasma by a solid-phase immunoassay (Enzyme Amplified Sensitivity
Immunoassay; Biosource; Flerus, Belgium). The method uses a blend of
monoclonal antibodies directed against different epitopes of TNF-
in
order to avoid hyperspecificity and to provide high
sensitivity.18
Intra-assay and interassay coefficients of
variation were 8% and 10%.
All results were corrected for hemodilution according to Cardigan et al19 by measuring total proteins in all the plasma samples and multiplying the various parameters at each time point by the ratio of protein levels at baseline to the levels at that time point. The neutrophil count at the various time points was done using a cell counter (TOA Sysmex NE-8000; TOA Medical Electronics; Carson, CA) and corrected for hemodilution.
Statistical Analysis
Descriptive statistics are reported as median and
interquartile range because of the skewed distribution of the variables
investigated, except for the relationship between MAP and bradykinin
concentrations (Fig 1
). The behaviors of the variables at the various times were assessed by
Friedmans nonparametric analysis of variance; in the case of a
significant result, pairwise comparisons with baseline time were made
by the nonparametric test of Wilcoxon-Mann-Whitney for paired data.
Bonferronis correction was done for multiple comparisons.
Nonparametric Spearmans correlation coefficient was calculated to
assess relationships between the variables at the various times.
|
| Results |
|---|
|
|
|---|
|
|
During ECC, there was no correlation between changes in bradykinin and in the other parameters. At the end of ECC, there was an inverse correlation between bradykinin levels and the weaning time (the time during which the pulmonary circulation is progressively restored while ECC continues) [r = - 0.636; p = 0.003; Fig 3 ]. Interestingly, at the same time (end of ECC), bradykinin levels were also inversely correlated with the polymorphonuclear blood cell count (neutrophils) [Fig 4 ]. Neutrophil counts were directly correlated with the weaning time (r = 0.565; p = 0.008). None of the other parameters were correlated with bradykinin levels at the end of ECC. The only correlation between the various parameters at any time was the strong one between prothrombin fragment F1 + 2 and TAT complexes (r = 0.801; p < 0.001).
|
|
|
| Discussion |
|---|
|
|
|---|
In the present study, we measured plasma bradykinin specifically after liquid-phase extraction and high-performance liquid chromatography separation of kinins using a method10 that had previously detected increased plasma bradykinin concentrations during attacks of angioedema. This method is based on careful sample handling and uses efficient inhibitors of bradykinin-generating and bradykinin-degrading enzymes. Baseline bradykinin levels in our patients with coronary artery disease are in the low picomolar range, thus comparable to those of healthy control subjects. From the beginning of CPB, the plasma bradykinin level rises. It is increased fourfold at the end of CPB and returns to normal by 24 h later. Thus, the hypothesis of increased plasma bradykinin concentrations during CPB was verified. The increase could be due to enhanced activation of the contact system as previously described in clinical CPB.26 In our patients undergoing CPB, FXIIa was indeed increased in parallel with bradykinin (Table 1) , but no direct correlation was found between individual changes in bradykinin and FXIIa levels. HK cleavage was not enhanced enough for immunoblotting to show an increase in HK breakdown products in plasma (Fig 2) , in contrast with overactive states of the kallikrein system, like hereditary angioedema.27 Nevertheless, a smaller degree of activation of the kallikrein-kinin system, beyond the sensitivity of Western blot analysis, cannot be excluded because HK has a micromolar plasma concentration but bradykinin has only a picomolar plasma concentration. Cleavage of a small fraction of HK (below the detection limit of Western blotting) might generate the few femtomols of bradykinin clearly detected by the bradykinin assay. Another reason for the increase in plasma bradykinin during CPB could be reduced degradation. Bradykinin is efficiently degraded in the pulmonary vascular bed by kininase II, ie, ACE. In animal experiments,28 bradykinin disappeared mostly in a single passage through the pulmonary circulation, and a clearance rate of 95% was extrapolated from bradykinin infusions in humans.11 CPB completely bypasses the pulmonary circulation, so bradykinin may well accumulate. In agreement with this reduced metabolism of bradykinin is our observation that plasma bradykinin levels at the end of CPB were inversely correlated with the weaning time, ie, the time it took for the pulmonary circulation to be gradually restored while CPB was still in progress (Fig 3) . If CPB is terminated after a long weaning time, the lungs are perfused for long enough to metabolize a large amount of bradykinin. If the weaning time is short, therefore, the decrease in plasma bradykinin may be small. To verify this, in six patients undergoing CPB, we took paired samples from the right atrium and the aorta after declamping the aorta and restoring the pulmonary circulation while ECC was ongoing. Bradykinin plasma levels were indeed 60% lower after passage through the natural lungs. Thus, in humans too, endogenous bradykinin is largely eliminated in the pulmonary vascular tree. Another reason for nondegradation of circulating bradykinin during CPB could be the absence of endothelium and thus of ACE in the tubings of the extracorporeal circuit. This was in fact found to cause artifactually increased angiotensin I levels in instrumented dogs.29 For bradykinin, this artifact has never been demonstrated, but our results are in full agreement with the idea of nondegradation, since CPB patients plasma bradykinin levels more than doubled from the entrance to the exit of the artificial circuit (Table 2) . Whatever the reason, however, for the increased plasma levels of bradykinin in tubings (lack of endothelium or activation of the contact system), the impressive 60% extraction of bradykinin across the natural lungs leads us to assume that the exclusion of the lungs during CPB is the most important reason for the increase in plasma bradykinin.
At the end of CPB, blood neutrophil counts were inversely related to plasma bradykinin. This would be compatible with bradykinin-mediated extravasation of neutrophils. Trafficking of neutrophils between the intravascular and extravascular compartment may be induced by the bradykinin-activation platform on the neutrophil surface,30 enhancing vasopermeation. This mechanism could contribute to the sequestration of neutrophils in the lungs at the end of bypass, also produced by activation of the fifth component of the complement system.7
Bradykinin is well-known as a vasodilator and hypotensive peptide. However, many other factors can induce hypotension during CPB, and in this condition the BP is maintained by the continuous intervention of the anesthetist. The drop in MAP observed before ECC was not associated with an increase of bradykinin and appeared after the induction of anesthesia, which is known to have a hypotensive effect. During CPB, if we exclude the seven patients who received vasoactive agents, the inverse correlation between bradykinin plasma levels and MAP suggests an important role of bradykinin. The inverse correlation between plasma bradykinin and the systemic vascular resistance further supports the hemodynamic effect of bradykinin during CPB.
Bradykinin levels were not correlated with the parameters of activation of coagulation, fibrinolysis, complement, and cytokines. The activation of coagulation during CPB is well-known and requires systemic heparin therapy to prevent gross clotting. However, the exact mechanism of activation is not completely understood, and both contact and tissue factor pathways may be involved.31 Our data show an increase of markers of thrombin generation (prothrombin fragment F1 + 2 and TAT complexes) not correlated with markers of contact system activation (cleaved HK and FXIIa), supporting the view of Boisclair et al31 that contact activation by exposure to foreign surfaces is not the only procoagulant stimulus during CPB. The increase of PAP complexes confirms the activation of fibrinolysis, already described in many studies32 33 34 and associated with blood loss during CPB.35 The significant reduction of PAP complexes we observed 24 h after CPB could be due to the fibrinolytic shutdown described in the postoperative period.36 The increases in C3a and TNF are in agreement with the activation of complement7 and cytokines37 during CPB. We did not find any correlations between the various parameters, except for prothrombin fragment F1 + 2 and TAT complexes, which were strongly correlated (r = 0.801; p < 0.001) as expressions of thrombin generation.38
In conclusion, our data demonstrate a progressive increase of plasma bradykinin levels during CPB, at least partially due to reduced elimination by the lungs. This increase, which may partially contribute to hypotension, is not correlated with activation of the contact system, coagulation, fibrinolysis, complement, or TNF. The effectiveness of the recently developed bradykinin-receptor antagonists in preventing complications of CPB remains to be established.
| Acknowledgements |
|---|
| Footnotes |
|---|
This study was partially supported by "Progetto Finalizzato IRCCS, Ospedale Maggiore, Milano" and by the Cardiovascular Research Foundation, Lausanne, Switzerland.
Received for publication July 21, 2000. Accepted for publication May 16, 2001.
| References |
|---|
|
|
|---|
2-antiplasmin-inhibitor complexes during cardiopulmonary bypass. Blood 73,468-471This article has been cited by other articles:
![]() |
A. D. Maslow, G. Stearns, P. Batula, C. S. Schwartz, J. Gough, and A. K. Singh The hemodynamic effects of methylene blue when administered at the onset of cardiopulmonary bypass. Anesth. Analg., July 1, 2006; 103(1): 2 - 8. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. McKeown and A. E. Epstein Future Directions: American College of Chest Physicians Guidelines for the Prevention and Management of Postoperative Atrial Fibrillation After Cardiac Surgery Chest, August 1, 2005; 128(2_suppl): 61S - 64S. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Toraman, S. Evrenkaya, M. Yuce, O. Turek, N. Aksoy, H. Karabulut, O. Demirhisar, and C. Alhan Highly positive intraoperative fluid balance during cardiac surgery is associated with adverse outcome Perfusion, March 1, 2004; 19(2): 85 - 91. [Abstract] [PDF] |
||||
![]() |
M. Pretorius, L. J. Murphey, J. A. McFarlane, D. E. Vaughan, and N. J. Brown Angiotensin-Converting Enzyme Inhibition Alters the Fibrinolytic Response to Cardiopulmonary Bypass Circulation, December 23, 2003; 108(25): 3079 - 3083. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |