| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2004;77:214-219
© 2004 The Society of Thoracic Surgeons
a Department of Immunology and Transfusion Medicine, Trondheim, Norway
b Department of Anesthesia and Intensive Care, St. Elisabeth Heart Center, University Hospital of Trondheim, Trondheim, Norway
c Department of Laboratory Medicine, Children's and Women's Health, Norwegian University of Science and Technology, Trondheim, Norway
Accepted for publication June 13, 2003.
* Address reprint requests to Dr Videm, Department of Immunology and Transfusion Medicine, St. Olav's Hospital, Trondheim, Norway N-7006.
e-mail: vibeke.videm{at}medisin.ntnu.no
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
METHODS: Blood samples from 21 patients undergoing combined valve and coronary artery bypass surgery were collected before anesthesia (T1), after re-transfusion of blood from the heart-lung machine (T2), and on the first postoperative morning (T3). Concentrations of soluble markers were determined using sandwich enzyme-linked immunoadsorbent assay for sICAM-1, sVCAM-1, and sE-selectin. The sera were also used to stimulate human umbilical vein endothelial cells (HUVEC) in culture for 6 hours, in which activation was measured using cell enzyme immunoassay for mICAM-1 and mVCAM-1.
RESULTS: The concentrations of sICAM-1 and sVCAM-1 increased during both measurement intervals (p < 0.05). The sICAM-1 T1 was 311.0 ng/mL (range, 271.0 to 350.7 ng/mL); the sICAM-1 T2 was 341.6 ng/mL (range, 322.0 to 422.0 ng/mL), and the sICAM-1 T3 was 400.2 ng/mL (range, 348.0 to 556.4 ng/mL; the sVCAM-1 T1 was 607.5 ng/mL (range, 497.8 to 813.8 ng/mL), the sVCAM-1 T2 was 755.3 ng/mL (range, 660.6 to 834.4 ng/mL), and the sVCAM-1 T3 was 1149.0 ng/mL (946.0 to 1406.0 ng/mL); whereas the sE-selectin increased from T1 to T3 (p < 0.01). Both the mICAM-1 (p < 0.002) and the mVCAM-1 (p < 0.005) increased on the human umbilical vein endothelial cells in culture after stimulation with the patient sera. The amounts of soluble markers in vivo were not correlated with the degree of endothelial activation in vitro, but were correlated with various operative variables including age, medication, and time of aortic cross-clamping.
CONCLUSIONS: Endothelial cells were activated during cardiopulmonary bypass. The soluble adhesion molecules sICAM-1, sVCAM-1, and sE-selectin displayed different kinetics, rendering it difficult to determine a simple expression for the degree of endothelial cell activation. Clinically, sVCAM-1 seemed to be the best-suited marker for endothelial cell activation, because it was only associated with aortic cross-clamping and heparin and protamine doses, and it also showed the largest numerical changes.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
Activation of endothelial cells and the subsequent increased surface expression of cell adhesion molecules are two of the fundamental steps in an inflammatory reaction involving adhesion and transmigration of leukocytes across the endothelium. The differential expression of adhesion molecules is a regulatory mechanism in this process. On the endothelium, E-selectin, ICAM-1 and VCAM-1 are adhesion molecules of special interest, because they are considered to be potential markers for endothelial activation [2].
E-selectin (CD62E) is a glycoprotein expressed only on endothelial cells after activation by inflammatory cytokines or endotoxin [3]. A soluble form (sE-selectin) is found in blood, and elevated levels of sE-selectin in serum have been reported in a variety of pathologic conditions including disseminated intravascular coagulation [4]. It may be anticipated that sE-selectin would suppress leukocyte migration by competing with E-selectin expressed on endothelial cells, but sE-selectin may actually activate neutrophils and act as a proinflammatory agent [5].
The ICAM-1 (CD54) is a glycoprotein belonging to the immunoglobulin superfamily, existing either as a transmembrane (mICAM-1) or soluble (sICAM-1) protein. The mICAM-1 is expressed among others, as endothelial cells, lymphocytes, monocytes, and eosinophils. Although constitutively expressed, mICAM-1 is upregulated by cytokines [6]. The sICAM-1 is most likely formed by proteolytic cleavage of mICAM-1, because synthesis from an alternatively spliced message has not been found yet [6]. Elevated levels of sICAM-1 appear to be associated with inflammatory conditions. However, with inflammatory conditions in which the ligands LFA-1 (CD11a/CD18) and Mac-1 (CD11b/CD18) are activated, binding and clearance of sICAM-1 may be enhanced. Thus a reciprocal relationship between sICAM-1 levels and inflammation is also possible [6].
The VCAM-1 (CD106) is also a glycoprotein belonging to the immunoglobulin superfamily. Cells expressing VCAM-1 include endothelial cells, neurons, smooth muscle cells, fibroblasts, and macrophages, but expression on endothelial cells dominates [7]. The endothelial expression of VCAM-1 can be stimulated by cytokines and requires de novo mRNA and protein synthesis [7]. The exact mechanism by which the soluble form sVCAM-1 is generated is unknown. It may involve both proteolytic processing and alternate splicing [8].
It is necessary to quantify the inflammatory response when working to improve the biocompatibility of CPB. It is generally held that the amount of soluble adhesion molecules reflects the degree of endothelial activation. The aim of this preliminary study was to examine the kinetics of soluble markers (sICAM-1, sVCAM-1, and sE-selectin) for endothelial cell activation in patients undergoing open heart surgery. Furthermore, we investigated if the markers were related to the degree of endothelial activation in human umbilical vein endothelial cell (HUVEC) cultures stimulated with patient sera, assuming that the HUVEC would reflect the endothelial mechanisms and reactions in vivo [9]. In addition, we evaluated the clinical relevance of the soluble markers.
|
Patients and methods |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
10 µg/kg). Cardiopulmonary bypass was performed using a Medtronic Affinity NTTR oxygenator (Medtronic Inc, Minneapolis, MN) with standard tubing and arterial line filters (Cobe Sentry with PrimeGard, 21 µm pore size [Cobe Cardiovascular Inc, Arvada, CO]). A venous reservoir (Medtronic Affinity CVR with 30 µm depth filter) was used (Medtronic Inc). The oxygenator was Trillium surface coated (Medtronic Inc). The extracorporal circuit was primed with Ringer's lactate containing 1 mg cefalotin (Keflin, Lilly, IN) and 7500 IE heparin (Nycomed Pharma, Asker, Norway). Before CPB, heparin (3 mg/kg body weight) was administered intravenously, and additional heparin was given if needed to maintain an activated clotting time of more than 480 seconds. Cardiotomy suction and a nonpulsative roller pump (Stöckert Instruments GmbH, Munich, Germany) was used in all patients. Cold St. Thomas's cardioplegia with added procaine hydrochloride and cold blood cardioplegia were used in addition to local cooling and moderate general hypothermia (32°C) [12]. Mixed venous oxygen saturation was obtained before rewarming and used as an indicator of tissue oxygenation. After CPB, protamine sulfate was administered based on activated clotting time and a dose-response curve made from the pre-heparin activated clotting time and the activated clotting time response to the first heparin administration. Catecholamines were given at the discretion of the attending physician. Concentrated red cells were administered if hematocrit was less than 20% during CPB and 25% after CPB, but advanced age or circulatory instability could trigger transfusion at higher hematocrits if considered indicated by the physician.
Blood drained from the thoracic cavity was continuously collected and retransfused through a standard transfusion filter (Cobe CTR filtered cardiotomy reservoir and PALL transfusion filter, 40 µm pore size [Cobe Cardiovascular Inc]). Autotransfusion was stopped when there was no substantial bleeding. The patients were extubated as soon as they were cooperative and showed stable hemodynamics and adequate oxygenation.
Blood samples were drawn at three times: before anesthesia (T1), after retransfusion of blood from the heart-lung machine (T2), and on the first postoperative morning (T3). After centrifugation, serum, or ethylenediaminete-traacetic acid to anticoagulated plasma was kept at -70°C until analyzed or used in cell culture.
Cell culture
Human umbilical vein endothelial cells were isolated and cultured as described by Jaffe and colleagues [13] with slight modifications. Primary cultures were grown to semiconfluent monolayers (
5 days) before stimulation with undiluted patient serum for 6 hours at 37°C.
Quantification of adhesion molecules
Enzyme-linked immunosorbent assays (R&D Systems, Minneapolis, MN) were performed in duplicates to measure sICAM-1, sVCAM-1, and sE-selectin in the patient sera. The normal ranges of these soluble adhesion molecules were (mean value and standard deviation) 211 ng/mL (range, 115 to 306 ng/mL) for sICAM-1 (n = 131); 553 ng/mL (range, 395 to 714 ng/mL) for sVCAM-1 (n = 105); and 43.3 ng/mL (range, 29.1 to 63.4 ng/mL) for sE-selectin (n = 130). Due to massive hemodilution, concentrations were corrected according to the method of van Beaumont, which considers the relationship between concomitant changes in hemoglobin and plasma volume [14]. The degree of activation in HUVEC cultures was measured using cell enzyme immunoassay for mICAM-1 and mVCAM-1. In this technique, antibodies against mICAM-1 or mVCAM-1 or a negative control antibody were added. After incubation the cultures were washed and an enzyme-conjugated goat antimouse (Dako, Glostrup, Denmark) secondary antibody was added. O-phenylenediamine (Sigma, St. Louis, MO) was used as substrate, and the optical density (absorbance) was read in a spectrophotometer (Elx800, BioTekInstruments, Winooski, VT).
Statistical methods
Nonparametric statistical analyses were performed due to non-normal distribution of variables. Data are given as medians with 95% confidence intervals. Friedman's test with Conover's method for multiple comparisons was used to compare measurements at T1, T2, and T3 [15]. Because of the different kinetics among the patients, the increases between T1 and maximal expression independent of time were also calculated. Wilcoxon's signed-rank test was used to assess differences between such pair wise observations. Spearman's rank correlation coefficient was computed to assess associations between continuous variables. Multivariate linear regression was used to investigate the effect of independent variables upon the concentration of soluble adhesion molecules. The independent variables included sex, age, body mass index, body surface area, aortic cross-clamping time, CPB time, autotransfusion volume, heparin dose, and protamine dose. Because of the small size of the study group, only a few important intraoperative variables were pre-selected for inclusion in the regression analyses. In addition, the demographic variables were included because they could introduce a bias. If necessary, variables were transformed (logarithmic or rank transformation) to achieve an appropriate model fit. The p values below 0.05 were considered statistically significant. The statistical packages Minitab (Minitab Inc, State College, PA) and SPSS 10.0 (SPSS Inc, Chicago, IL) were used.
|
Results |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
|
|
|
|
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
Endothelial activation in HUVEC cultures
Human umbilical vein endothelial cells are large venous endothelial cells present in the umbilical cords of newborns, and their characteristics may not be comparable with endothelial cells in the adult vasculature. However, comparative studies of endothelial cells from the human vena saphena and from the femoral artery proved HUVEC to be a relevant model for the study of adhesion molecules [16], especially if primary cultures are used.
After incubation of HUVEC with sera from the patients undergoing CPB, mICAM-1 showed an increase after surgery followed by a decrease the next morning. In the case of mVCAM-1, there was considerable variation in the time course of the responses, which may reflect the fact that this adhesion molecule requires de novo mRNA and protein synthesis. The concentrations of mVCAM-1 were nevertheless increased from baseline to maximal expression, indicating that the endothelial cells had significantly upregulated the expression for mVCAM-1.
The observed kinetics of the adhesion molecules indicate that CPB resulted in secretion of compounds capable of inducing endothelial activation and enhanced expression of adhesion molecules. The data are consistent with previous knowledge about the kinetics of many pro-inflammatory mediators that are found in their highest concentrations immediately after CPB. Candidates for such endothelial cell activators in serum are plenty. Endotoxin increases in concentration during CPB, and is a well-known activator of endothelial cells. Although the surgical equipment and medication may contain endotoxin, the most important source of this compound is believed to be a leakage through the gut due to increased permeability [17]. Furthermore, products from the plasma cascade systems contribute to activation, together with cytokines and mediators from neutrophil granulocytes and platelets [1].
Soluble adhesion molecules in vivo
The markers displayed different kinetics as shown in Figure 1. The concentration of sICAM-1 increased mostly between T1 and T2. The sVCAM-1 displayed a slower response, because the increase from T1 to T2 constituted a small share of the total increase. This may in part be explained by the rate of transcription making the time delay for protein synthesis greater for sVCAM-1 than for sICAM-1. The sE-selectin concentrations also increased during the operation, but the numerically low concentrations made this marker difficult to evaluate any further.
Little is known about the significance of the kinetics and the sources of the soluble adhesion molecules. They are assumed to be released from the surface of activated endothelial cells and leukocytes by means of proteolytic cleavage. The sE-selectin is specific for activated endothelial cells, and therefore can be regarded as evidence for endothelial activation. The sICAM-1 can originate from several types of cells, and alone, it can not indicate endothelial activation. The sVCAM-1 is also secreted from several cell types, but endothelial cells are believed to be the major source. Therefore, elevated levels of sVCAM-1 give a strong indication of endothelial activation.
The relationship between adhesion molecules on the cellular surface and the level of soluble forms is also still poorly understood. It is often assumed that the level of soluble adhesion molecules in serum or increased mRNA-expression is parallel to the concentration of adhesion molecules on the cell surface. There are several weaknesses to such a conclusion, because elevated serum levels can also reflect the rate of synthesis, the rate of proteolytic cleavage, or reduced removal of the molecules in their soluble forms. The expression in cell cultures did not correlate with the in vivo data, which may partly be explained by the extremely small surface in a culture well compared with the body's vasculature. Moreover, there are other turn-over mechanisms in the intact organism, making it difficult to determine a simple expression for endothelial activation. It is possible that the HUVEC model reflects production only, and not the breakdown of the adhesion molecules.
Previous studies have demonstrated diverse effects of CPB upon the levels of circulating adhesion molecules [18, 19], but the majority show an increase in concentration during and after CPB. Nevertheless, it can be questioned whether the increase is caused by CPB or by the substantial operative trauma. In a comparative study, cardiac operations were associated with increased plasma levels of soluble adhesion molecules in contrast with complex, long-lasting abdominal or lung operations [20]. It was concluded that activation of pro-inflammatory cascades, ischemia and reperfusion, and microcirculatory dysfunction appeared to be the most likely reasons for the observed differences between the two groups. Another potential explanation for the increase in soluble markers is through the re-transfusion of shed mediastinal blood, but this was less important as shown by regression modeling.
Several studies have demonstrated that the release of adhesion molecules to the circulation is strongly correlated with the degree of endothelial trauma in experimental settings, and also multiple organ dysfunction and other severe conditions such as systemic inflammatory response syndrome and sepsis [21]. However, none of these studies were focused on clarifying the suitability of the adhesion molecules as markers for endothelial activation during CPB.
Correlation between soluble markers and operative variables
The concentrations of sICAM-1 in both T2 and T3 were negatively correlated with the patients' ages, which ranged from 50 to 82 years. The correlation was stronger in T2. This may indicate that aged endothelium can not respond as quickly as before and consequently the sICAM-1-release appeared later. Other mechanisms associated with the proteolytic cleavage of mICAM-1 may also be involved. In one study, it was concluded that elderly patients are more prone than younger patients to pronounced activation or even damage of the endothelium [22]. However, this was based on a 5-day observation period and a wider age range (< 50 years and > 70), and therefore this is not comparable with the present study.
The concentration of sVCAM-1 in T2 was positively correlated with heparin and negatively correlated with protamine. Heparin and protamine complexes are activators of the complement system, which may be responsible. Another possible explanation is that free circulating heparin acts pro-inflammatory, because heparin in the relevant doses primes neutrophils for activation [23], which may be a link to endothelial activation. The concentration of sVCAM-1 in T3 was positively correlated with the duration of aortic cross-clamping. Hypoxia and ischemia during aortic cross-clamping can lead to endothelial activation and subsequent expression of adhesion molecules [24]. We may speculate that ischemia is the most relevant trigger for endothelial activation and counts for a stronger effect than that of foreign surface contact, rewarming and so forth. This information, together with the lack of association with re-transfused blood, indicates that the triggers for endothelial activation occur much earlier in the course of the operation. Our data show that the sVCAM-1 changes were numerically larger than those of other markers, which may be an advantage in clinical settings.
The sE-selectin concentrations were positively correlated with the heparin dose in T2, and heparin and protamine doses in T3. This may also be related to activation of the complement system. Because E-selectin is internalized as well as secreted [25], it may take some time before the changes in circulating sE-selectin become detectable. Age was negatively correlated with the concentration of sE-selectin in T3. This further strengthens the hypothesis of age-related changes in endothelial function as described for sICAM-1.
The present pilot study was too small to allow evaluation of the markers' possible relations to postoperative complications. For a definite conclusion of the markers' relevance in biocompatibility studies, larger follow-up studies are needed. However, the present study does provide some interesting indications. The relationship of sICAM-1 to the patients' age is not advantageous and tells nothing about the response to the heart-lung machine. The concentrations of sE-selectin were low and displayed minor changes, thereby reducing sensitivity. In addition, sE-selectin was correlated with age. On the other hand, the sVCAM-1 concentrations were only correlated with CPB-related variables such as heparin and protamine doses and the time of aortic cross-clamping. A previous study concluded that sVCAM-1 could possibly function as an early marker for endothelial activation in connection with systemic inflammation during open heart surgery [26]. The present data are in accordance with those results. Because sVCAM-1 is also more endothelial cell-specific, sVCAM-1 stands out as a potentially important marker in biocompatibility studies of CPB.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. Foglieni, F. Maisano, L. Dreas, A. Giazzon, G. Ruotolo, E. Ferrero, L. Li Volsi, S. Coli, G. Sinagra, B. Zingone, et al. Mild inflammatory activation of mammary arteries in patients with acute coronary syndromes Am J Physiol Heart Circ Physiol, June 1, 2008; 294(6): H2831 - H2837. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Farneti, S Sbrana, D Spiller, A. Cerillo, F Santarelli, D Di Dario, P. Del Sarto, and M Glauber Reduction of blood coagulation and monocyte-platelet interaction following the use of a minimal extracorporeal circulation system (Synergy(R)) in coronary artery bypass grafting (CABG) Perfusion, January 1, 2008; 23(1): 49 - 56. [Abstract] [PDF]
|
|
|
|
 |
|
 C. Schneider and T. Volk Endothelial Ca2+ signal transduction is altered by postoperative serum from patients undergoing coronary surgery with cardiopulmonary bypass. Anesth. Analg., June 1, 2006; 102(6): 1630 - 1637. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F.-X. Schmid, B. Floerchinger, N. K. Vudattu, G. Eissner, M. Haubitz, E. Holler, R. Andreesen, and D. E Birnbaum Direct evidence of endothelial injury during cardiopulmonary bypass by demonstration of circulating endothelial cells Perfusion, May 1, 2006; 21(3): 133 - 137. [Abstract] [PDF]
|
|
|
|
 |
|
 Y. Misawa Markers of Endothelial Activation During Cardiopulmonary Bypass Ann. Thorac. Surg., March 1, 2005; 79(3): 1096 - 1096. [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 |
