| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1996;62:1926-1934
© 1996 The Society of Thoracic Surgeons
Biomaterial Research Unit (INSERM U.306) and Department of Anesthesia, Centre Hospitalier Universitaire de Bordeaux, Bordeaux, France, and Department of Anesthesia, Montreal Heart Institute, University of Montreal, Montreal, Quebec, Canada
Abstract
Background. Performance of a majority of cardiac surgical procedures requires the use of extracorporeal circulation. Contact of the patients' blood with the nonendothelial surface of the cardiopulmonary bypass circuit is responsible for several, potentially harmful systemic reactions.
Methods. The patients' response to extracorporeal circulation is reviewed briefly. The interactions between patient and circuit are discussed not only as they relate to blood-material contact, but also from a mechanical and rheologic standpoint. The theoretic benefits of the newer, more hemocompatible materials are presented, along with a review of published clinical experience with heparinized cardiopulmonary bypass circuits.
Results. The response to extracorporeal circulation extends far beyond a simple derangement of hemostasis. This inflammatory response is strongly influenced by the rheologic design of the circuit and by the physical and chemical properties of the surface. Heparinized circuits decrease inflammation, but the clinical benefits of this reduction remain unclear, except for extended cardiopulmonary support. The safe use of these circuits requires full heparinization and does not reduce allogeneic transfusions.
Conclusions. Clinicians are still in the search of the ideal material and the ideal extracorporeal circuit design. Newer, heparinized materials offer real but limited clinical benefits.
During open heart operations, patients' blood must be continuously processed using extracorporeal circuits fitted with pumps and suitable active components (eg, specific filters, oxygenators). Contact of blood proteins and cells with the nonendothelial surface of the cardiopulmonary bypass (CPB) circuit activates the five plasma proteolytic systems (coagulation, fibrinolysis, complement cascade, kallikrein-kinin, and contact systems) and at least three cellular elements (leukocytes, platelets, endothelial cells). Contrary to the normal situation whereby these mechanisms are localized and intended to promote wound healing, the intensity of the stimulus encountered during CPB results in a nonlocalized, potentially harmful, systemic reaction, the so-called whole-body inflammatory response. Our objective is to deceive the response systems so that they do not recognize the foreign surface of the CPB circuit. To this end, the materials composing such circuits, intended to be in contact with blood for variable periods of time, must be as hemocompatible as possible [1].
The general term "hemocompatibility" refers to those properties that allow CPB circuits to maintain contact with flowing blood without causing any adverse reactions, without releasing any leachable components, and without suffering any alteration.
A number of properties will define the biocompatibility of a material. These are:
These attributes depend not only on the surface characteristics of a given material, but also on extrinsic conditions such as sites of canulation, duration of contact with blood, and local hemodynamic status (eg, pulsatile versus nonpulsatile flow, length and diameter of tubes).
The Inflammatory Response to Cardiopulmonary Bypass
In response to trauma, for example, after being exposed to a foreign surface such as that of the CPB circuit, blood tissue will initiate the production of proteolytic substances. This will result in the production of thrombin, plasmin, and C3a and C5a by activation of the coagulation, fibrinolysis, and complement cascades respectively; kallikrein results from activation of the kinin-kallikrein system. Although the production of proteolytic substances is initiated by proteins present in plasma (humoral phase), its amplification results mainly from the contribution of the cellular elements of blood (cellular phase). The proteolytic substances thus generated exert their effects on membrane receptors and on other circulating proteins. Action of these cascades must be circumscribed, in view of avoiding exaggerated and undesirable systemic effects. To this end, potent inhibitors exert a local control of activated factors. Overpowering or exhaustion of these restrictive mechanisms results in the systemic manifestation of cascade activation such as disseminated intravascular coagulation, primary fibrinolysis, or the systemic inflammatory syndrome.
Activation of proteins known as contact phase proteins is responsible for the first response of blood tissue to exposure to CPB material. This leads to the coagulation of blood or plasma via the intrinsic pathway of coagulation. In the presence of foreign surfaces, Hageman factor (factor XII) forms activated factor XII (factor XIIa), which promotes the formation of kallikrein from prekallikrein. In the presence of high-molecular-weight kininogen, kallikrein helps in the formation of factor XIIa [2]. The generation of factor XIIa is a self-amplified process. Along a classic cascade of reactions, the prothrombinase complex (factor Xa, phospholipids, and factor V) is formed and is able to transform the circulating prothrombin (factor II) into active thrombin (factor IIa). In addition to the production of fibrin, thrombin triggers a set of self-amplified processes (via factors Va and VIIIa) and activates factor XIII and protein C. Thrombin is also a potent platelet agonist.
Simultaneously with the activation of the contact phase of coagulation, which gives rise to thrombin, platelet adhesion and aggregation result in the release of adenosine diphosphate, platelet factor 4, thromboxane A2, and factor V (release reaction), and in the expression of receptors and phospholipids at the platelet surface (flip-flop reaction). The surface of the platelet is thus the milieu where are concentrated the humoral components of coagulation.
As mentioned previously, exposure of blood to foreign surfaces activates the alternate pathway of the complement system [3, 4] whereby C3 convertase is formed directly from the C3 component. Only a very small part of this component needs to be cleaved to provide circulating C3b fragments. The latter may bind covalently to several kinds of surfaces. This is the starting point for the formation of a complex with protein B.
At this stage the bound protein B can be cleaved by a serine esterase to give the alternate C3 convertase C3b and Bb, and to release the fragment Ba. The C3b-Bb complex, stabilized by properdin, is then available to cleave more and more C3, the main consequences of which are the release of the anaphylatoxic fragment C3a on the one hand, and the formation of the alternate C5 convertase on the other. In the case of nonactivating surfaces, bound C3b binds to protein H instead of protein B, preventing C3 convertase generation. The C5 convertase provokes the cleavage of the C5 component, which releases the anaphylatoxin C5a, whereas the fragment C5b may form a ternary complex with C6 and C7 components; this complex is able to penetrate the bilipid layer of cell membranes. This insertion is the first step of the elaboration process of a cytolytic complex. This complex is made of C8 component and several molecules of C9 component, added to the preexistent ternary complex. It should be recalled here that the components C5 to C9 may attach to platelets and amplify thrombin-induced platelet release and aggregation.
The cellular reaction of blood to biomaterials is not limited to platelets. Fibrin degradation products, as well as fibrinopeptides, which are generated by the action of thrombin on fibrinogen, are chemotactic for polymorphonuclear leukocytes, whereas the anaphylatoxins C3a and C5a are chemotactic for both polymorphonuclear leukocytes and monocytes. These cells will be attracted toward the site where the complement system is activated, or where thrombogenic phenomena or fibrinolysis are taking place. They will express on their membrane an adhesive protein known as Mo1, which promotes their aggregation and their adhesion to endothelial cells. The Mo1 antigen is supported by one of the two macromolecular chains that are constitutive of the so-called CR3 receptor. This kind of receptor, present on the membrane of polymorphonuclear leukocytes or monocytes, recognizes specifically the C3bi fragment of the C3 component. This fragment comes from the cleavage of the C3b fragment by protein I, when C3b is bound to a nonactivating surface (Fig 1
).
|
Endothelial cells and monocytes may also secrete platelet-activating factor, which promotes and amplifies the inflammatory response when cells are stimulated by adenosine diphosphate, histamine, and bradykinin. Whether the phenomenon is due to effects from factor XIIa, or to those of a plasminogen-activating factor, plasmin is generated, and this enzyme is able to cleave the C3 component.
In response to the effects of various activating substances, endothelial cells may behave as procoagulant agents, a function opposite to their normal physiologic behavior. The expression of thrombomodulin on their surface may be interrupted. Thrombomodulin is a natural scavenger for thrombin and may neutralize adenosine diphosphate also. Adenosine diphosphate activates endothelial cells but, at the same time, promotes platelet aggregation.
Ultimately, other factors may help explain the interindividual variability of the response to the CPB circuit. These include reserves of restrictive factors, baseline or intraoperative modulation of immune reactivity, and the contribution of other phenomena occurring during CPB but not directly related to the blood/CPB surface interaction, such as nonpulsatile flow and blood/wound interactions.
Mechanical and Rheologic Considerations
Blood is a nonnewtonian fluid, composed of a suspension of cells occupying a significant fraction of the total volume of a nonhomogeneous fluid (42%).
The flow of a fluid in a tubular structure is described by Poiseuille's law. Unfortunately, blood velocity profiles cannot be determined according to Poiseuille's law, essentially for two reasons. First, blood flow is either pulsatile or nonpulsatile locally, and second, it may be turbulent (Poiseuille's law can be applied only to laminar flow).
Blood velocity profiles are of paramount importance because they determine the wall shear rate, the main parameter of the stress suffered by cells when they make contact with the walls of vessels, heart cavities, or synthetic surfaces. Wall shear rate increases with flow rate and, because the flow rate may oscillate between extreme values, the wall shear rate may do the same, which is hazardous to cells.
Interactions between blood elements and the surface they encounter must be differentiated from those occurring between the elements themselves in the core of the flow. Cell-to-cell interactions are influenced by local flow conditions, whereas cell-to-wall interactions are influenced by the local shear rate. Local shear rate is related directly to local flow conditions, which is, in turn, determined by the geometry and the mechanical properties of the conduit, the characteristics of the upstream pressure wave, and the viscosity of the fluid.
The first type of interaction that occurs when blood elements reach the wall result from the competition between diffusive and convective transports. The former tends to increase the concentration of a given subset of cells that tend to react with special sites on the wall and with products generated by related reactions. The diluting effect of blood flow can be prevented by abnormal flow situations (poor design of conduits or pathologic alterations of the circulatory tree). Such situations induce appearance of spaces characterized by the presence of eddies and vortices, spaces where a poor turnover of blood is observed. In these spaces where thrombogenic products are seldom washed out and concentrate, the potential for blood cell activation processes is much more favorable. Reactions at the blood-wall interface as well as local flow reactions may occur only if they are thermodynamically favored, and the actual occurrence of any particular reaction among competitive processes depends upon its rate constant.
Surface Characteristics and Blood-Material Interactions
Biomaterials may be compact or porous. Compact materials with smooth surfaces are preferred as blood cells will encounter fewer opportunities to be injured by asperities or morphologic singularities. The mechanical trauma to cells depends on the size of the morphologic singularities. The mechanical trauma to cells depends on the size of the morphologic abnormalities coupled to local flow conditions (eg, shear stress, turbulence, vortices). When surfaces feature peaks or valleys (average height or depth = 9 µm), the number of platelets adhering to polyvinyl chloride surfaces can be multiplied by 3, compared with polished surfaces. Closed porosities of variable size may be used to modify the density of biomaterials to optimize their mechanical properties. For such porous materials, however, the surface morphology must fulfill the same requirements as those of compact materials. Generally, such materials cannot be exposed directly to flowing blood for long periods of time without anticoagulation. For example, a comparison between two different pediatric membrane oxygenators indicates that a flat sheet membrane oxygenator has a higher complement activity than a hollow-fiber membrane oxygenator. This is probably due to the relatively larger blood-surface contacting area of the flat-sheet membrane oxygenator [6].
The surface tension, or surface free energy, is a parameter that describes the residual binding capacity of atoms (or of groups of atoms) at the exposed surface of the material. This binding capacity may be uniform (nonoxidized metals) or it may be the result of several components related to the various types of atoms or atom groups present on surface. These may include ionic sites, hydrophobic sites, polar sites, or hydrogen atom donors or acceptors. Many material surfaces appear as a mosaic-like structure. In addition to the mechanical and rheologic characteristics described previously, the nature of the potential sites of interaction and the biochemical microtopography of these different sites will determine the interactions between biomaterials and blood during CPB.
Blood cells and vessel walls are charged negatively, the corresponding isoelectric point lying at a pH between 4.8 and 5 when the voltage difference is measured between the endothelial layer and blood [7]. Negatively charged surfaces give rise in the presence of an electrolyte solution to a double electric layer responsible for the recorded potential value 
0 (Fig 2a) between the surface (eg, intimal surface) and the solution (eg, blood).
|
0 decreases linearly down to a value equal to 0/2.3. The corresponding distance (
) to the wall is known as the Debye length. At distances greater than , the potential decreases exponentially to zero, the decaying potential being characterized by the so-called zeta (
) potential (Fig 2b
). The distance from the wall, corresponding to , defines the shear plane and depends upon the flow conditions. For an electrolyte solution such as blood, increases linearly with the flow rate. The vessel wall being negatively charged, it repels the negatively charged platelets and helps prevent thrombogenic phenomena. It has been shown experimentally that on positively charged surfaces (>200 mV/normal hydrogen electrode) thrombosis occurs systematically, whereas the thrombosis never occurred on surfaces with a negative potential (<0 mV/normal hydrogen electrode).
The electrical criterion cannot be the unique parameter of nonthrombogenicity. The superficial distribution of the charged sites plays an important role as far as plasma protein adsorption is concerned. The nature of the adsorbed proteins may trigger the coagulation cascade, even if the bloodstream is exposed to a negatively charged surface.
At first, in situations where blood flow conditions have no influence on the initiation of interactive phenomena between blood and biomaterials, materials show an affinity for water that depends on their wettability, in turn related to their surface tension. Polar materials and materials bearing ionic sites are obviously the most wettable. Hydrophobic materials do not interact with water, but instead promote its intermolecular organization, as in the case of oil or paraffins. The contribution of water to the entropy of the system is thus decreased. This first step may occur simultaneously with the interaction of free ions in solution with the ionic sites present on the surface of the material or be immediately followed by the interaction of small solutes with the material. The ions and small solutes contribute to a local alteration of the water organization at the interfacial region.
The second step concerns the adsorption of proteins and other biological macromolecules at the blood-material interface [9]. These adsorption phenomena are controlled by the nature and the decrease in available binding sites of the surface, and by the state of the surface after the previous step. Macromolecules and proteins are characterized by a distribution of binding sites at their surface. If there is a good fit between this distribution and the one exposed by the material surface, proteins may adsorb without any conformational change. Among available macromolecules, those for which this fit is the best have the greatest probability to adsorb (Fig 3). These adsorption phenomena may give macromolecules an opportunity to be activated and initiate the contact phase of coagulation.
|
Low-stress, shear-induced hemolysis appears to be sensitive to wall material chemistry, to be proportional to device-to-volume ratio (surface/volume ratio), to have slow kinetics, and to be controlled by fluid shear rate, while being essentially independent of shear stress [11].
Despite the importance of the physical and chemical properties of a surface on the interaction between blood and a biomaterial, this interaction will also depend considerably on the local flow patterns of the blood near the surface. It has been shown that plasma protein adsorption, which occurs as soon as blood comes into contact with a surface, is clearly influenced by the local shear stress [12].
The presence of high concentrations of activated coagulation proteins secondary to these biochemcial reactions and flow characteristics may favor the activation of platelets and leukocytes. These two types of blood cells are able to undergo activation via collision processes, favored by turbulent flow or particular conditions such as those created by vortices [13].
Platelet and fibrin deposition are influenced by plasma and cellular factors (for example, a high concentration of red blood cells increases the radial movement of platelets and enhances their diffuseness by several orders of magnitude) and by shear rate. Weiss and associates [14] have shown that platelet deposition on the subendothelium is at a minimum at a shear rate of 50 s-1 while fibrin deposition is at a maximum. At low shear rates (250 s-1), the fibrin deposition is independent of platelet density and integrity, and at higher shear rate (650 s-1), platelets must possess all their physiologic properties to promote fibrin deposition. At much higher shear rates (2,600 s-1) fibrin deposition decreases, despite increasing platelet deposition.
Conception and Elaboration of Hemocompatible Materials
As thrombogenesis may result from several interconnected pathways, researchers have failed in their attempt to prepare synthetic materials that do not give rise to the generation of thrombin when they are placed in contact with blood, whether the generation of thrombin results from an activation of the contact phase in the intrinsic pathway of coagulation or from more indirect mechanisms involving platelets or leukocytes and the extrinsic pathway of coagulation. Thus, there is considerable interest in preparing materials with a surface of thrombin-inhibiting, antiplatelet or profibrinolytic character. Some materials may combine these different properties. Because of their biological properties, all of these materials can be termed "bioactive materials."
Materials With Thrombin-Inhibiting Properties
Heparinized materials derive their activity from the known ability of heparin to bind to antithrombin III and to dramatically increase kinetics of the inhibition of thrombin by antithrombin III. Heparin can be attached to blood-contacting materials according to various techniques. Two main classes of materials can be distinguished: those that release heparin, and those to which heparin is irreversibly bound.
In the case of materials that release heparin, the heparin molecule can either be mixed with the polymeric blend, which is then extruded [15], or mixed with polyvinyl–alcohol-based hydrogels [16] and used to coat the inner surface of tubes. Taking into account its polyanionic character, heparin can also be ionically bound to surfaces bearing a suitable distribution of cationic sites [17]. These principles have been largely applied, although slight differences do exist between the numerous procedures proposed. For the resulting materials, heparin activity mainly comes from the heparin they release because of the relative instability of ionic bonds.
To manufacture materials to which heparin is irreversibly bound, heparin is covalently bound to the macromolecular backbone of the material. Most procedures take advantage of both the reactive hydroxyl groups of heparin [6] and the presence of identical or different reactive groups on the material of interest. Whichever coupling agent is used, the glycosaminoglycan molecule becomes associated to the material by more than one covalent bond, the consequence of which is a severely decreased conformational mobility of the heparin molecule and of its catalytic activity. Other procedures bypass this drawback by binding heparin molecules with a very limited number of bonds [6].
Heparin-like materials may be used instead of heparin. The concept underlying these materials is based on the hypothesis that only a few structural units among those constituting the heparin macromolecules are necessary for the expression by the latter of its affinity for antithrombin III and for the resulting effect on the inhibition of thrombin. In this respect, the -NH-SO3- group expressed by some of the osamine units plays an outstanding role. This postulate has been proposed to improve the hemocompatibility of various materials by chemical modification of their surface, involving the introduction of sulfonate groups and of sulfonamide groups (SO2-NH-CH2-R, where R is an amino acid residue). A wise choice of the latter as well as an optimal density and balance of sulfonamide and sulfonate groups gives the material a strong affinity for antithrombin III and its resulting, desirable heparin-like catalytic activity [18].
Because the affinity of heparin for antithrombin III is mediated by a particular pentasaccharide unit, attempts have been made to bind either this pentasaccharide or heparin fragments containing this pentasaccharide to biomaterials. This heparin-like agent is, at present, too expensive to be available in current practice. An alternate proposal involves the use of hirudin, a peptidic antithrombinic agent. Hirudin can now be prepared on a large scale through biotechnological processes, and hirudin-bearing materials correspond to a strategy that is economically feasible.
Materials With Proactivating Properties of Fibrinolysis
Activation of plasminogen may be mediated either by a tissue-related factor (tissue plasminogen activator), liberated by endothelial cells, or by other activators such as urokinase. It is possible to bind such activators to blood-contacting materials to endow them with the ability to promote fibrinolysis of thrombi. However, the mechanism by which these activators function may be incompatible with their binding to a foreign surface (tissue plasminogen activator, for instance, must adhere to the fibrin network to be active).
Materials That Inhibit Platelet Aggregation
Prostacyclin (prostaglandin I2) cannot be used because it is highly labile. Salicylic acid, some nonsteroidal antiinflammatory drugs, and other, more stable prostaglandin derivatives can be linked to macromolecular materials from which they can be released progressively. These compounds may also be bound by macromolecular coupling chains long enough to offer as much freedom as possible to the pharmacologic agent, to allow it to express optimal activity [19].
Design of the Ideal Material
The ideal material should not activate the complement system and should promote the adsorption of plasma proteins to prevent platelet adhesion. Several authors have identified a correlation between the behavior of materials when they contact blood (as far as protein adsorption and cell activation and adhesion phenomena are concerned) and the free energy of their surface [20]. According to their hypothesis, surfaces with a given, critical surface tension (25 ergs•cm-2) would be the most hemocompatible. In effect, surface tension is a parameter that expresses quantitatively an average residual binding capacity but that says nothing about the finer chemical or physical characteristics of the surface. Nothing is known about the nature of the sites responsible for this binding capacity or about their distribution at a molecular level, ie, on the scale of the factors that determine the nature of the interactions. Thus, there may exist many surfaces for which the critical surface tension approximates 25 ergs•cm-2, but very few for which the surface chemistry will match adequately that of biomolecules, the adsorption of which is required for an optimal interaction of blood with the material.
The ideal biomaterial should bind albumin and prevent the adsorption of fibrinogen, factor XII, and high-molecular-weight kininogen, suppress the activation of C3, and decrease hemolysis during CPB. All the components of the CPB circuit could benefit from such a nonthrombogenic surface. These are the challenges we face when designing new biomaterials, in view of improving outcomes after CPB.
Clinical Experience With Heparinized Cardiopulmonary Bypass Circuits
Heparinized Circuits and Hemostasis
The activation of coagulation, as characterized by the generation of thrombin, progresses during CPB despite the administration of enormous doses of heparin systemically [21]. This generation of thrombin is not reduced by the use of heparin-coated circuits with full systemic heparinization [22, 23], or may even be augmented when compared with nonheparinized circuits [24]. When reduced systemic heparinization and heparin-coated circuits are employed, the production of prothombin fragments (F1+2) is accelerated [24, 25], reflecting the intense generation of thrombin. In this context, thrombin generation is increased when compared with that observed when full systemic heparinization and conventional circuits are used [24, 25]. Because heparin-coated circuits induce a thrombotic stimulus similar to that of conventional circuits, reducing systemic heparinization carries a risk of increasing fibrin microemboli and intravascular and CPB circuit clotting [26, 27].
Thus, the generation of thrombin increases progressively during clinical CPB, especially after release of the aortic clamp and reperfusion of the heart and lungs [23], despite the use of heparinized circuits. However, it should be noted that these increases are modest compared with the major increases of F1+2 observed 2 hours after operation regardless of which type of CPB circuit was used [25].
The impact on the inhibition of thrombin by heparin-coated circuits has also been studied. Fibrin degradation products [28] and fibrinopeptide A concentrations [28, 29] remain unchanged when heparin-coated circuits are used. As for F1+2, levels of circulating thrombin-antithrombin complexes may be either elevated [24, 25] similar [22], or reduced [29, 30] when heparinized circuits are compared with conventional materials. The postoperative elevation of thrombin-antithrombin complexes is similar to that seen for F1+2. Heparinized circuits adsorb significantly more antithrombin III, thus sequestering more thrombin but no less fibrinogen, factor XII, and von Willebrand factor than standard circuits [22]. Fewer leukocytes, fewer platelets, and less fibrin adhere to the arterial filter of heparinized circuits [31].
In summary, the effect of heparin-coated circuits on the generation/inhibition of thrombin appears to be minor, especially in the presence of full systemic heparinization. The reduced antithrombin III activity observed commonly during clinical CPB could help explain, in part, these nonconclusive, or even paradoxic, results.
The impact of heparinized circuits on the platelet count and on the liberation of ß-thromboglobulin remains controversial, but it is probably minor [22–25, 28] when compared with other factors, including the administration of protamine. In the study by Videm and associates [32], a further decrease in the platelet count (by approximately 60,000 x 106/L) was observed after the administration of protamine whether heparinized or conventional circuits had been used. Gorman and colleagues [22] reported a smaller decrease in adenosine diphosphate induced platelet aggregation in the presence of heparin-coated circuits. However, when conventional circuits are employed in conjunction with an aggressive anticoagulation protocol, it is also possible to preserve platelet function after CPB [33]. Thus, as for the generation/inhibition of thrombin, the beneficial effects of heparin-coated circuits on platelet count and function appear to be marginal.
Heparinized Circuits and the Transfusion of Allogeneic Blood Products
In the presence of full systemic heparinization, the use of coated versus conventional CPB circuits is associated with similar postoperative blood losses, and the requirements for allogeneic blood products remain unchanged [22, 23, 28, 30, 34].
When heparinized circuits and reduced heparin doses were compared with conventional circuits with full heparinization, decreased postoperative bleeding was reported in four studies [25, 27, 35, 36] and decreased allogeneic blood product transfusions in three [27, 35, 36]. However, these observations have not been confirmed by several other trials [24, 32, 37–39]. Decreases in postoperative bleeding averaged 60 mL [25], 249 mL [27], 1,514 mL [36], and 507 mL/m2 [35]. However, the mechanisms by which postoperative bleeding or transfusions were decreased must be discussed. First, a hypercoagulable state secondary to the decreased systemic heparin dose may be involved [26]. Second, considerable (and inappropriate) postoperative bleeding in the control group, as in the study by von Segesser and associates [36], makes it easier to demonstrate the beneficial effect of any intervention [40]. Third, the demonstration that transfusion of red blood cells, plasma, and platelets decreased threefold, fourfold, and fivefold, respectively, when mediastinal chest drainage decreased only 249 mL casts doubt on the appropriateness of the transfusion protocol followed [27]. Finally, as discussed previously, the reduced protamine dosage may also have played an important role in the beneficial effects observed. Whatever the predominant mechanism may be, it remains unclear at present if the benefit of reduced bleeding is worth the possible catastrophic risks of intravascular clotting associated with reduced systemic heparinization [26].
A case of CPB (with heparinized material) for aortic valve replacement without pharmacologic anticoagulation has been reported [41]. It should be noted, however, that this patient with end-stage hepatic disease presented with disorders of coagulation, thrombocytopenia, and decreased platelet aggregation and required transfusion of hemostatic blood products to control bleeding after the operation. Cardiopulmonary bypass with heparin-coated circuits but without some form of systemic anticoagulation, whether it be pharmacologic or acquired through disease, has never been described in humans. Systemic anticoagulation remains essential to prevent clotting of noncirculating blood, as may be found in the coronary arteries or in cardiac chambers at different times during the operation.
Extracorporeal membrane oxygenation is a good model of prolonged exposure to an imperfectly biocompatible material. Since the introduction of heparin-coated circuits for extracorporeal membrane oxygenation, several groups have reported decreased bleeding and transfusion requirements in those patients [42, 43]. Confirmation of these initial observations should document the benefits of heparinized perfusion systems when prolonged exposure to the circuit is anticipated.
Heparinized Circuits and Inflammation
Complement activation in association with CPB [44, 45] and after heparin neutralization with protamine [46, 47] has been described repeatedly. The reduction of complement activation is the most consistent benefit reported with heparin-coated CPB circuits [32, 34, 38, 48, 49], and is mediated through inhibition of the amplification of the alternate complement pathway [50] and the subsequent reduction of C5b-9. Perfusion with heparinized circuits significantly reduces C3a generation after the administration of protamine [51], a time when complement activation is maximal [49]. Heparin-coated circuits are associated with a reduced generation of cytokines interleukin-6 [49, 52] and interleukin-8 [52, 53] during the first 24 hours postoperatively.
The benefits of heparin coating notwithstanding, it must be realized that circuit design plays a major role in determining the extent of complement and granulocyte activation associated with CPB [37, 54]. Thus, equivalent complement activation may be achieved with circuits from different manufacturers, whether they are heparin coated or not, depending on the quality of the underlying design.
Increased levels of C3a 3 hours after CPB have been associated with a greater incidence of cardiac, pulmonary, renal, and hemostatic dysfunction [55]. Peak concentrations of C5b-9 and C3a correlated with increased [32] or abnormal bleeding after CPB [55]. The clinical benefits of reducing inflammation after extracorporeal circulation remain somewhat speculative, especially in the low-risk patient, but may be real in the high-risk patient, especially with regard to preservation of pulmonary integrity [34, 49, 56].
Conclusions
The contact of blood with the extracorporeal circuit can be compared to an extensive wound, exposing the interior "milieu." In response to this aggression, the organism initiates multiple wound healing processes, most often referred to as the "inflammatory response" to CPB. Given the importance of the "wound," this inflammatory response cannot be circumscribed (as it is in the normal wound healing process), becomes systemic, and is initiated with the activation of the coagulation and complement cascades. Research in the field of hemocompatibility aims to deceive the organism as to the true nature of the CPB circuit, and progress has been accomplished in several areas (eg, by the reduction of porosity and improvement of the rheologic characteristics of the circuit).
Although heparin-coated circuits have improved biocompatibility, they are still imperfect. They decrease the inflammatory response to CPB, but thrombogenesis is not reduced and complete systemic heparinization should be maintained to avoid complications secondary to the activation of coagulation. Heparinized circuits may be of benefit in high-risk patients, in cases where prolonged CPB times are anticipated, or for extended cardiopulmonary support with devices such as the extracorporeal membrane oxygenator. The totally biocompatible extracorporeal circuit remains a challenge for biomedical engineers.
Footnotes
Address reprint requests to Dr Janvier, Département d'Anesthésie-Réanimation II, Hôpital Cardiologique, Groupe Hospitalier Sud, 33600 Pessac, France.
References
This article has been cited by other articles:
|
|
 |
|
  S. Rubin, A. N. Ali, O. N. Pages, and B. Baehrel How to replace an extracorporeal life support without interruption of the cardiopulmonary assistance Interact CardioVasc Thorac Surg, August 1, 2009; 9(2): 311 - 313. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. F. Padera Jr. and F. J. Schoen Pathology of Cardiac Surgery Card. Surg. Adult, January 1, 2008; 3(2008): 111 - 178. [Full Text]
|
|
|
|
 |
|
 S. G Raja and G. D Dreyfus Modulation of Systemic Inflammatory Response after Cardiac Surgery Asian Cardiovasc Thorac Ann, December 1, 2005; 13(4): 382 - 395. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Gunaydin Clinical significance of coated extracorporeal circuits: a review of novel technologies Perfusion, January 1, 2004; 19(1_suppl): S33 - S41. [Abstract] [PDF]
|
|
|
|
|
|
F. J. Schoen and R. F. Padera Jr. Cardiac Surgical Pathology Card. Surg. Adult, January 1, 2003; 2(2003): 119 - 185. [Full Text]
|
|
|
|
 |
|
 H. P. Wendel, N. Hauser, F. Briquet, and G. Ziemer Hemocompatibility of Medical Connectors with Biopassive or Bioactive Surface Coatings J Biomater Appl, July 1, 2002; 17(1): 5 - 17. [Abstract] [PDF]
|
|
|
|
 |
|
 K. Kottke-Marchant and S. Sapatnekar Hemostatic Abnormalities in Cardiopulmonary Bypass: Pathophysiologic and Transfusion Considerations Seminars in Cardiothoracic and Vascular Anesthesia, September 1, 2001; 5(3): 187 - 206. [Abstract] [PDF]
|
|
|
|
|
|
H. P Wendel, A. Philipp, N. Weber, D. E Birnbaum, and G. Ziemer Oxygenator thrombosis: worst case after development of an abnormal pressure gradient - incidence and pathway Perfusion, July 1, 2001; 16(4): 271 - 278. [Abstract] [PDF]
|
|
|
|
 |
|
 U. Lockowandt, A. Owall, and A. Franco-Cereceda Myocardial outflow of prostacyclin in relation to metabolic stress during off-pump coronary artery bypass grafting Ann. Thorac. Surg., July 1, 2000; 70(1): 206 - 211. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. P. Wendel and G. Ziemer Coating-techniques to improve the hemocompatibility of artificial devices used for extracorporeal circulation Eur J Cardiothorac Surg, September 1, 1999; 16(3): 342 - 350. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Reents, J. Babin-Ebell, M. R. Misoph, A. Schwarzkopf, and O. Elert Influence of different autotransfusion devices on the quality of salvaged blood Ann. Thorac. Surg., July 1, 1999; 68(1): 58 - 62. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. P Wendel, A. M Scheule, F. S Eckstein, and G. Ziemer Haemocompatibility of paediatric membrane oxygenators with heparin-coated surfaces Perfusion, January 1, 1999; 14(1): 21 - 28. [Abstract] [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 |
