|
|
||||||||
a Centre Hospitalier Régional Universitaire de Lille, Clinique de Chirurgie Cardiovasculaire, Lille, France
b Institut d'Hématologie-Transfusion, Lille, France
c Inserm ER193, Lille, France
d Université de Lille, Faculté de Médecine, Lille, France
Accepted for publication February 22, 2008.
* Address correspondence to Dr Fabre, Service de Chirurgie Cardiaque, Hôpital Cardiologique, Lille Cedex, 59037, France (Email: o-fabre{at}chru-lille.fr).
| Cardiothoracic anesthesiology:
The Annals of Thoracic Surgery CME Program is located online at http://cme.ctsnetjournals.org. To take the CME activity related to this article, you must have either an STS member or an individual non-member subscription to the journal.
|
| Abstract |
|---|
|
|
|---|
Methods: In 10 patients, we analyzed hemostasis activation markers and inflammatory mediators in blood collected in the pericardial cavity and in blood actively aspirated from the left ventricle without any contact with the pericardial cavity. In addition, the same variables were determined in blood sampled in the cardiopulmonary bypass circuit.
Results: Markers of tissue factor pathway activation and of thrombin generation, microparticles, free hemoglobin, interleukin 6, and tumor necrosis factor-
were significantly increased in pericardial samples as compared with the left ventricle and cardiopulmonary bypass circuit samples. All measured variables were similar between left ventricle and cardiopulmonary bypass samples, except free hemoglobin, interleukin 6, and microparticle levels, which were significantly higher in the left ventricle.
Conclusions: Blood contact with the pericardial cavity induces strong hemolysis, inflammatory mediator release, and coagulation activation, driven by tissue factor pathway activation. By contrast, suction forces applied to left ventricular blood poorly contribute to blood trauma and activation. Comparison of pericardial and left ventricular blood shows that contact with the pericardial cavity, and not suction forces, is the leading cause of blood activation. The specific trigger for blood trauma and activation present in the pericardial cavity remains to be identified.
Cardiac surgery with cardiopulmonary bypass (CPB) induces major hemostasis and inflammation activation, which are thought to participate to organ dysfunctions in the postoperative period [1, 2]. Despite strong anticoagulation by heparin, a significant rate of thrombin generation remains detectable, which is mainly mediated by the tissue factor (TF) pathway of blood coagulation and microparticle generation from activated leukocytes, red blood cells, and platelets [3–5]. A particularly strong activation process has been demonstrated in blood aspirated from the pericardial cavity [6–8] in association with high levels of inflammatory mediators and hemolysis [9–11], but the underlying activating mechanisms are unclear. Contact of blood with the pericardial cavity and surgical wound is considered as a leading cause [11, 12]. However, other triggers have been suggested to play a major role, such as exposure of blood to suction forces and air contact, although their role in TF-mediated coagulation activation and inflammation is poorly known [13, 14].
During CPB, blood originating from lungs is continuously and actively aspirated from the left ventricle to obtain a continuously dry operative field. This blood is submitted to suction forces and variable air contact, and not to contact with the pericardial cavity. To separately analyze the effects of active suction and those of contact with the pericardial cavity, we compared the activation state of blood aspirated from the left ventricle to that of blood directly sampled in the pericardial cavity without active suction.
| Patients and Methods |
|---|
|
|
|---|
Cardiopulmonary Bypass Procedure
The extracorporeal system device included a standard nonpulsatile roller pump (Jostra; Maquet, Rastatt, Germany), a hard-shell venous reservoir, and a hollow-fiber membrane oxygenator with heat exchanger (Dideco, Mirandola, Italy) according to the practice in our center. The CPB circuit included three noncoated lines: one for systemic circulation, one for suction in the pericardial cavity and the surgical field, and one for active venting of the left ventricle. Systemic venous blood was drained by gravity in the reservoir. The CPB circuit was primed with 1,800 mL of Ringer's lactate solution, 125 mL of 4.2% sodium bicarbonate, and 10,000 IU of heparin. Before starting CPB, heparin was infused intravenously (300 IU/kg). Systemic heparinemia was maintained greater than 3 IU/mL throughout CPB. Antifibrinolytic therapy consisted of a bolus of tranexamic acid (50 mg/kg) before the skin incision, followed by continuous infusion (10 mg/kg per hour) until the end of surgery. Cardiopulmonary bypass was established between the right atrium and the ascending aorta. The core temperature was allowed to lower to 34°C. Cardiac arrest was induced and maintained by intermittent antegrade potassium blood cardioplegia at 11°C (for 1.5 minutes every 30 minutes). During aortic cross-clamping, the venting of the left ventricle was achieved by insertion of a catheter in the aortic root, which is connected to an independent roller pump. During CPB, blood aspirated from the pericardial cavity was collected in a cell-saving device (Cell Saver 5, Haemonetics Corp, Braintree, MA). Before the end of CPB, this blood was centrifuged, washed, and returned to the venous reservoir.
Blood Sampling
Blood sampling was realized in each patient after induction of anesthesia (baseline) and 30 minutes after the onset of CPB (immediately before the first reinjection of cardioplegia) from (1) the pericardial cavity, (2) the line for left ventricular suction, and (3) the venous line of the CPB. Blood from the pericardial cavity and the CPB circuit was sampled through gentle aspiration with a syringe, whereas blood from the left ventricle was sampled through the surgical aspiration line. No patient received blood products before blood sampling.
Blood was collected into EDTA for hematocrit and blood cell numeration, into 129 mmol/L sodium citrate anticoagulant (Becton-Dickinson, Franklin Lakes, NJ; ratio of blood to anticoagulant was 9:1 vol/vol) for platelet and coagulation analysis, and into dry tubes for cytokine determination. Citrated platelet-poor plasma was obtained by centrifugation for 10 minutes at 3,000 g. All plasma and serum samples were snap-frozen and stored at –80°C until assayed.
Plasma Assays
Tissue factor pathway activation was evaluated through dosage of total plasma TF antigen (soluble and microparticle-bound) using an enzyme-linked immunosorbent assay technique (American Diagnostica Inc, Stamford, CT) and of activated factor VII (FVIIa) using a recombinant TF protein that does not promote FVII activation according to Morrissey's procedure (Staclot VIIa-rTF; Diagnostica Stago, Asnières sur Seine, France) [15]. Thrombin-antithrombin (TAT) complexes, interleukin 6 (IL-6), and tumor necrosis factor
(TNF-
) were determined by enzyme-linked immunosorbent assay techniques using commercially available kits (TAT: CSL Behring, King of Prussia, PA; IL-6 and TNF-
: R&D Systems, Minneapolis, MN) for thrombin generation and inflammation exploration, respectively. Plasma-free hemoglobin was assessed by spectrophotometric measurement using an iEMS reader MF spectrophotometer (LabSystems, Cergy Pontoise, France) at 561, 576, and 597 nm.
Platelet Assays
Flow-cytometric analysis of platelet activation was performed on citrated blood using an XL flow cytometer (Beckman Coulter, Fullerton, CA) with the use of the Platelet GP kit for platelet membrane glycoprotein quantitation (Biocytex, Marseille, France). Single color-flow cytometric analysis of P-selectin was performed without and with stimulation by thrombin receptor activator peptide. The number of P-selectin antigenic sites was determined by converting the fluorescence intensity into the corresponding number of sites per platelet based on a calibrated bead standard curve.
Microparticle Assay
Platelet-free plasma samples were obtained by subsequent centrifugation of citrated platelet-poor plasma at 13,000 g for 2 minutes at room temperature. Microparticles were then isolated through their expression of phosphatidyl serine by a solid-phase capture assay onto immobilized annexin V. Prothrombinase assay was used for the determination of the amount of captured procoagulant microparticles (nmol/L; phosphatidyl serine equivalent) as previously described [16].
Statistical Analysis
All results are expressed as the mean ± standard error of the mean. Overall comparison among the three groups was assessed using Kruskal-Wallis test. When significant differences were noted, comparison between two groups was assessed using Mann-Whitney U test for unpaired comparisons. Probability values less than 0.05 were considered significant.
| Results |
|---|
|
|
|---|
Hematocrit, platelet count, and blood cell counts were lower during CPB than at baseline but were not different among the sampling sites during CPB (data not shown). Baseline values (after correction for per CPB hemodilution) of each determined inflammatory and hemostasis variable were similar to those sampled from CPB, except for FVIIa which tended to decrease after the onset of CPB (Fig 1).
|
, and free hemoglobin levels were significantly higher in pericardial cavity blood than in CPB blood (p
0.005), and also significantly higher in pericardial cavity blood than in left ventricular blood (p
0.01). Microparticles, IL-6, and free hemoglobin levels were higher in left ventricular than in CPB blood (p
0.05). Plasma TF antigen was significantly different between pericardial cavity and CPB blood (p = 0.005) and tended to be higher in pericardial cavity than in left ventricular blood, and in left ventricular than in CPB blood, although the difference did not reach statistical significancy (Fig 1). After flow-cytometric analysis, platelet membrane P-selectin expression was similar at baseline in platelets from the three sampling sites. However, stimulation of thrombin receptor activator peptide induced enhanced expression of membrane P-selectin in platelets from left ventricular and CPB blood (p = 0.02), and not from pericardial cavity samples (Fig 2).
|
| Comment |
|---|
|
|
|---|
Strong hemolysis has already been described in pericardial blood [17], and was at first considered as a result of blood trauma induced by suction forces [13]. We confirm that suction forces induce mild hemolysis on left ventricular blood, but we clearly demonstrate that hemolysis and strong TF-mediated coagulation activation are present in pericardial blood before it was submitted to suction forces.
Noteworthy, platelet P-selectin expression was similar among all sampling sites. In addition, platelets collected from the pericardial cavity failed to express P-selectin after thrombin receptor activator peptide stimulation, compared with platelets aspirated from the left ventricle. This result suggests that suction forces do not induce significant platelet impairment, and that a specific trigger exhausts the potential for platelet activation in pericardial blood. Whether this strong impairment of platelet function contributes to postoperative bleeding remains to be investigated.
The occurrence of significant thrombin generation mediated by TF pathway activation is clearly established during CPB [18–21]. Tissue factor is the cell receptor and activator for factor VII, and the principal activator of physiologic coagulation [22]. The presence of both soluble and microparticle-associated TF has been demonstrated in pericardial blood and can induce early thrombin generation during the surgical procedure [5, 23, 24]. We confirm the presence of high levels of soluble TF, FVIIa, microparticles, and TAT complexes in pericardial blood. In this study, we did not address the issue of the relative role of soluble TF and microparticle-bound TF on FVIIa generation. However, it was recently demonstrated that microparticle-free soluble TF, bound to peripheral monocytes, is the most efficient activator of factor VII [24, 25]. By contrast, despite a slight increase of microparticles and TF, no increase of FVIIa ant TAT was detectable in blood aspirated from the left ventricle, suggesting that suction force–mediated cell trauma plays a minor role in thrombin generation and TF pathway activation.
Pericardial blood also contains high levels of inflammatory cytokines, as previously underlined [9, 11, 26]. These inflammatory mediators are of unknown origin, but they may be a trigger for leukocyte and endothelial activation. We found that IL-6 levels in left ventricular blood, although lower than in pericardial blood, were significantly increased as compared with CPB blood, confirming that nonventilated lungs are a source of inflammatory mediators, as previously described [1]. However, this contribution appears to be modest as compared with the amount of TNF-
and IL-6 released in pericardial blood.
Besides suction forces, contact with air has been advocated as a potent blood activator, prompting the use of closed circuits in some centers [14]. In our study, blood from the three sampling sites underwent contact with air: blood sampled from CPB had contact with air in the open venous reservoir, active suction of the left ventricle induces variable air contact depending on the importance of the negative pressure and of the flow rate applied by the occlusive pump, and blood shed in the pericardial cavity was in contact with air before sampling. We thus cannot exclude that air contact participated in blood activation. However, as pericardial blood was more strongly activated than blood from the other two sampling sites, one may conclude that air contact by itself is not the leading factor for pericardial blood activation.
This study confirms that pericardial blood is highly activated. Reinfusion of this blood has been demonstrated to be potentially deleterious, through increasing postoperative bleeding and the need for transfusion of blood products [12, 27], inducing deleterious hemodynamic effects [28], and increasing the risk of neurologic injury [29, 30]. These findings have prompted the proposition that blood aspirated from the pericardial cavity should be discarded or processed through a cell-saving device [31, 32]. In addition to these demonstrated notions, our study shows that, in coronary artery bypass grafting surgery, despite active suction and variable air contact, blood from the left ventricle is poorly activated, indicating that it can be routinely reinfused without deleterious consequences.
In summary, this study shows that blood trauma and activation are mainly driven by contact of blood with the pericardial cavity and not by suction forces. The results are hemolysis, generation of microparticles, TF-mediated activation of blood coagulation, and inflammatory cytokines release. This study adds further arguments for minimizing the volume of blood in contact with the pericardial wound, and discarding or processing of this blood fraction with a cell-saving device. Most importantly, identification of the triggering factor for hemolysis, TF activation, and inflammation mediator release in pericardial blood may raise other new strategies aimed at minimizing the deleterious effects of CPB during cardiac surgery.
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
Related Article
This article has been cited by other articles:
![]() |
F. De Somer Optimal Versus Suboptimal Perfusion During Cardiopulmonary Bypass and the Inflammatory Response Seminars in Cardiothoracic and Vascular Anesthesia, June 1, 2009; 13(2): 113 - 117. [Abstract] [PDF] |
||||
![]() |
M. E. Jessen Invited Commentary Ann. Thorac. Surg., August 1, 2008; 86(2): 541 - 542. [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 |