|
|
||||||||
J Thorac Cardiovasc Surg 1994;108:636-641
© 1994 Mosby, Inc.
CARDIOPULMONARY BYPASS, |
Paris, France
Received for publication Jan. 11, 1994. Accepted for publication May 10, 1994. Address for reprints: I. Philip, MD, Département d'Anesthésie et de Reanimation Chirurgicale, Hopital Bichat, 46 rue Henri-Huchard, 75878 Paris Cedex 18, France.
Abstract
To determine the cytokine release during normothermic cardiopulmonary bypass, we have measured plasmatic levels of tumor necrosis factor-
and interleukins-1ß, 6, and 8 in 10 patients during the first 24 hours after the start of bypass. Arterial blood samples were collected at intervals before, during, and after bypass. Interleukin-1ß was not detectable in the plasma, and traces of tumor necrosis factor-
were detected in only three patients at times independent of the cardiopulmonary bypass procedure. Circulating endotoxin remained undetectable. Plasma interleukin-6 and interleukin-8 rose significantly from 2 until 24 hours after the start of bypass (p < 0.05) and peaked respectively at 4 and 2 hours after the beginning of bypass (interleukin-6, 268.1 ± 131.43 pg/ml; interleukin-8, 370 ± 420 pg/ml; mean peak ± standard deviation). Peak values of interleukin-6 and interleukin-8 were correlated neither with the duration of aortic crossclamping or the bypass procedure nor with the hemodynamic parameters recorded at the same times. This study shows that normothermic cardiopulmonary bypass does not induce systemic release of tumor necrosis factor-
and interleukin-1ß. A local production of these cytokines cannot be excluded, because interleukin-6 and interleukin-8 are produced by stimulated macrophages and monocytes in response to tumor necrosis factor-
and interleukin-1ß. Our results, at normothermia, show a similar pattern of interleukin-6 and interleukin-8 release when compared with release during hypothermic cardiopulmonary bypass. Interleukin-8, an important chemotactic neutrophil factor, might play a role in reperfusion injuries observed in lungs and heart after cardiopulmonary bypass. (J THORAC CARDIOVASC SURG 1994;108:636-41)
Exposure of blood to extracorporeal artificial surfaces during cardiopulmonary bypass (CPB) induces an acute inflammatory response.
1 After cardiac surgery, this response is a combination of interaction of blood components with artificial surfaces, reperfusion injury, and activation by endotoxin.
2,3 The inflammatory process might be associated with morbidity caused by organ dysfunction observed after CPB. Different inflammatory pathways are activated after CPB, including the complement system,
4,5 the contact system, thepolymorphonuclear neutrophils,
6,7 and the mononuclear cells, leading to cytokine release.
2,8
Indeed, previous studies have reported different patterns of cytokine production after hypothermic CPB. Haeffner-Cavaillon and associates
9 demonstrated intracellular production of interleukin-1 (IL-1) in monocytes. Interestingly, no circulating IL-1ß has been found in the plasma of patients after hypothermic bypass.
8,10 Conflicting data are available for tumor necrosis factor-
(TNF
), because it remained undetectable in some studies,
9-13 although others have measured significant plasma values of TNF
after bypass.
14,15 TNF
and IL-1ß, when produced in excess, might induce deleterious effects, including endothelial dysfunction with enhanced vascular permeability, decreased systemic vascular resistances, and myocardial depression.
16-18
High levels of interleukin-6 (IL-6) have been reported after cardiac surgery,
2,8,11,12 but the clinical relevance remains unclear. Neutrophil activation described after CPB might be induced by interleukin-8 (IL-8), known to be a major neutrophil chemotactic factor.
19 Recent investigations have reported its pathophysiologic role in different clinical situations, particularly in reperfusion injury.
20 Three studies have found an increase in plasma IL-8 levels after hypothermic CPB.
11,21,22
Warm heart surgery, with warm blood cardioplegia and normothermic CPB, has been recently introduced. Warm blood cardioplegia seems to be an effective method of myocardial protection.
23,24 Normothermic CPB shortens CPB duration and minimizes platelet dysfunction.
25 Nevertheless, previous studies have reported lower systemic vascular resistance during normothermic CPB when compared with hypothermic CPB.
26,27 A greater inflammatory response in normothermia, with enhanced complement activation and IL-1 release, has been documented in vitro.
9 An enhanced cytokine release after normothermic CPB, mainly TNF
and IL-1ß, might partly explain hemodynamic differences between warm and cold surgery.
The aim of our prospective work was to measure circulating concentrations of TNF
, IL-1ß, IL-6, IL-8, and endotoxin in 10 patients undergoing normothermic CPB for cardiac surgery. Indeed, cytokine production has been studied only after hypothermic CPB, and to our knowledge no data are available concerning normothermic CPB.
PATIENTS AND METHODS
Patients
After approval by the local ethics committee, 10 patients were prospectively included for the study. They underwent a valve operation with normothermic CPB. Their ages ranged from 23 to 83 years (46.8 ± 16 years, mean ± standard deviation). No corticosteroids were administered before or during the study. Patients with suspected infectious disease were excluded. Details on patients and length of CPB are summarized in
Table I.
|
Serial blood samples (5 ml) were withdrawn from the radial artery catheter before sternotomy (T0); at the beginning of CPB (H0); and at intervals of 1, 2, 4, 6, 10, and 24 hours (H1 to H24) after the beginning of CPB. Blood samples were collected into sterile vacuum tubes with ethylenediaminetetraacetic acid and immediately centrifuged. An aliquot of plasma was obtained and stored at -70° C until cytokine and endotoxin assays were performed. Mean arterial pressure, cardiac index (assessed by thermodilution), systemic vascular resistances, and bladder temperature were recorded all along the study.
Enzyme immunoassays for IL-1ß, IL-6, IL-8, and TNF
The levels of plasma immunoreactive IL-1ß, IL-6, IL-8, and TNF
were determined by immunoenzymatic assays according to the manufacturer's procedure. IL-1ß, IL-6, IL-8, and TNF
enzyme-linked immunosorbent assays were from Amersham (Biotrak, Les Ulis, France). The detection limits were 3 pg/ml for IL-1ß and IL-6, 10 pg/ml for TNF
, and 18 pg/ml for IL-8. Results were not corrected for hemodilution.
Measurement of endotoxins
Endotoxin was determined using the Limulus amebocyte lysate assay from TechGen (Limusate, Les Ulis, France). The sensitivity of this assay was 0.06 EU/ml.
Statistical analysis
The data were analyzed by analysis of variance. Correlations between peak cytokine values and different parameters were assessed by Spearman's rank correlation. Significance was accepted when p was less than 0.05. Results are expressed as mean ± standard deviation.
RESULTS
Patients
Table I shows the clinical characteristics and operative procedures for the 10 patients. Mean CPB and aortic clamping durations were 113.2 ± 17.6 and 88.7 ± 16.5 minutes, respectively. Temperature was 36.7° ± 0.4° C during CPB and reached a maximal value of 37.7° ± 0.6° C at H10. All of the patients recovered uneventfully, except one patient who required surgical hemostasis for major bleeding at H12.
Cytokines and endotoxin
IL-1ß and TNF
. Il-1ß was never detectable in our 10 patients until H24. TNF
was undetectable except in three patients. When detectable, plasma TNF
concentrations were low (<100 pg/ml) and no correlation was found with operative times.
IL-6. Fig. 1 shows changes in plasma concentration of IL-6. Plasma IL-6 was detectable in one patient at T0 (12 pg/ml). Plasma levels remained unchanged after 1 hour of CPB, raised significantly at H2, and peaked 4 hours after the start of CPB, remaining above baseline values 24 hours after CPB. The mean peak value at H4 was 268.1 ± 131.43 pg/ml. IL-6 plasma levels rose significantly in all patients.
|
, which remained undetectable.
|
Correlation between changes in cytokine levels and hemodynamic data
There was no correlation betweenpeak levels of IL-6 or IL-8 and CPB or aortic clamping durations. Fig. 3 shows mean arterial pressure and systemic vascular resistances during the study. No correlation was found between either IL-6 or IL-8 concentrations and mean arterial pressure, cardiac output, or systemic vascular resistances.
|
By contrast to the numerous data on cytokine release after hypothermic CPB, to our knowledge no data are available after normothermic CPB. This lack led us to measure, in 10 patients undergoing normothermic CPB, circulating levels of the main cytokines that could be involved in postoperative organ failure. We found an increase in plasma IL-6 and IL-8 concentrations after normothermic CPB, whereas IL-1ß and TNF
remained undetectable. Warm heart surgery was introduced a few years ago, but few prospective studies have been reported concerning the innocuousness of this technique.
23,24 Some previous studies concerning warm heart surgery reported lower systemic vascular resistance with this technique when compared with hypothermia, the "gold standard."
26,27 Warm heart surgery could increase vasopressor requirements.
26 Greater blood viscosity might contribute to increased systemic vascular resistance during hypothermia, but the possibility of an enhanced cytokine production after normothermic CPB has not been explored. Indeed, the temperature has a major influence on the different inflammatory pathways activated during CPB. Neutrophilia during CPB is prevented by hypothermia but reappears quickly when the body temperature is restored to 36° C.
28 In vitro, hypothermia reduces complement activation and inhibits IL-1ß production by monocytes.
9 Furthermore, clinical investigations have reported increased levels of cytokine only after rewarming of the patients.
9,11,14,21,29
In hypothermia, conflicting data are available concerning circulating levels of TNF
or IL-1ß. In many studies, plasma IL-1ß and TNF
have not been detected after hypothermic CPB,
8,10,12,13,21 except insome patients of two studies.
14,15 Some possibilities might explain the discrepancies. During CPB, variable levels of endotoxin have been reported,
14,30,31 with enhanced plasma levels of TNF.
32 Endotoxin was not detected in the plasma of our patients. The short half-life of TNF
and IL-1ß and differences between methods of cytokine measurement might also explain discrepancies between different studies. Bioassays detect biologic activity, measured in vitro on specific cells. By contrast, immunoassays measure serum antigens. The latter may detect complexes of cytokine and cytokine-soluble receptor or cytokine inhibitor as free cytokine and thus provide false-positive levels, when compared with bioassays.
33 Furthermore, the absence of detectable circulating cytokines does not exclude a likely local production by activated cells.
34
In our study, we did not find significant levels of circulating TNF
or IL-1ß after normothermic CPB. This argues against a deleterious effect of normothermic CPB with an exacerbated production of these two cytokines, which might induce vasodilation and contribute to postoperative complications. Moreover, systemic vascular resistance did not decrease significantly in our patients. Only two patients required injection of phenylephrine (250 and 500 µg) to maintain systemic perfusion pressure above 50 mm Hg. Anesthetic management with high doses of fentanyl and neither halothane nor isoflurane might explain hemodynamic stability. Moreover, in our study myocardial protection was obtained with intermittent cold blood cardioplegia and not with continuous warm cardioplegia.
IL-6, known to have pleiotropic functions, is one of the key mediators in the acute phase response.
35 In the present study, plasma IL-6 levels increased significantly 2 hours (H2) after the start of CPB, peaked at H4, and remained elevated at H24. The peak was earlier than peaks observed after moderate hypothermic CPB, and thus normothermic CPB seems to quicken the inflammatory response. Nevertheless, although constantly observed after cardiac operations, a rise in plasma IL-6 levels is not specific to CPB because it also occurs after major noncardiac operations.
36-40
Neutrophil activation has been documented after CPB, with the release of the peroxidation products,
7 myeloperoxidase,
6 lactoferrin, and elastase.
6,10 The role of neutrophil activation in lung and myocardial injuries has been well documented in experimental or human investigations.
2,6,7,41-43 This led us to measure IL-8, a potent neutrophil chemotactic and activating factor.
19 Our results show an early increase in plasma IL-8 levels, with peak levels 2 hours after the beginning of CPB. The fact that peak values were not correlated with the length of aortic clamping or bypass might be due to the homogeneous durations of CPB in our patients. Recently, a rise in plasma IL-8 levels has been reported after hypothermic CPB.
11,21,22 Even if comparisons must be done with caution, it appears that normothermic CPB induces a pattern of IL-8 production similar to that observed with hypothermic CPB.
11,21,22
IL-8 production does not seem to be specific to cardiac surgery, but rather to situations of ischemia-reperfusion or anoxia-hyperoxia injuries.
21,22,44 Lung reperfusion injury in rabbits has been shown to be prevented by a monoclonal antibody against IL-8.
20 These studies emphasized the major role that IL-8 might play in the reperfusion injury described after cardiac operations. Thus we are now investigating the clinical relevance of IL-8 secretion on pulmonary function and neutrophil activation during normothermic CPB.
The cells involved in the production of IL-8 after cardiac surgery remain to be determined. Pulmonary production by alveolar macrophages might be a potent source of IL-8. Furthermore, high levels of this cytokine have been detected in the bronchoalveolar lavage fluid of patients after CPB.
22,45 A local production of TNF
and IL-1ß, even if undetectable in the circulation, could induce IL-8 synthesis and secretion.
In conclusion, our study shows that normothermic CPB is associated with IL-6 and IL-8 production, which is similar to that previously observed after hypothermic CPB. Plasma levels of TNF
and IL-1ß remain undetectable in our patients after CPB, results that are consistent with previous studies concerning hypothermic CPB. Our results suggest that normothermic CPB per se does not induce an overproduction of cytokines when compared with hypothermic CPB.
Footnotes
From Département d'Anesthésie et de Réanimation Chirurgicale,a Unité INSERM U408,b Laboratoire de Biochimie A,c and Service de Chirurgie Cardio-vasculaire,d Centre hospitalo-universitaire Bichat, Paris, France. ![]()
References
after elective cardiopulmonary bypass. Perfusion 1993;8:233-8.
after elective cardiopulmonary bypass. Perfusion 1991;6:51-4.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |