Biomaterials 25 (2004) 5681–5703 Review Biomaterial-associated thrombosis: roles of coagulation factors, complement, platelets and leukocytes Maud B. Gorbet, Michael V. Sefton* Department of Chemical Engineering and Applied Chemistry, Institute of Biomaterials and Biomedical Engineering, University of Toronto, 4 Taddle Creek Road, Room 407D, Toronto, Ont., Canada M5S 3G9 ARTICLE IN PRESS 3. Complement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5686 3.1. Classical pathway . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5686 3.2. Alternative pathway . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5687 3.3. Regulatory molecules of complement activation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5688 3.4. Interactions of complement and coagulation cascade . . . . . . . . . . . . . . . . . . . . . . . . . 5688 3.5. Complement activation and biomaterials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5688 *Correspondin E-mail addres 0142-9612/$ - see doi:10.1016/j.bio Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5682 2. Coagulation cascade . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5683 2.1. The intrinsic pathway . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5683 2.2. The extrinsic pathway . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5684 2.3. Physiologic inhibitors of coagulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5684 2.4. Biomaterials and coagulation pathways . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5685 2.5. Anticoagulants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5686 these components in thrombosis and with biomaterials and cardiovascular devices. We also briefly highlight a few features of biomaterial-associated thrombosis that are not often considered in the biomaterials literature: * The importance of tissue factor and the extrinsic coagulation system. * Complement activation as a prelude to platelet activation and its role in thrombosis. * The role of leukocytes in thrombin formation. * The differing time scales of these contributions. r 2004 Elsevier Ltd. All rights reserved. Keywords: Leukocytes; Tissue Factor; CD11b; Platelets; Biomaterials; Complement activation; Thrombosis; Coagulation Abstract Our failure to produce truly non-thrombogenic materials may reflect a failure to fully understand the mechanisms of biomaterial- associated thrombosis. The community has focused on minimizing coagulation or minimizing platelet adhesion and activation. We have infrequently considered the interactions between the two although we are generally familiar with these interactions. However, we have rarely considered in the context of biomaterial-associated thrombosis the other major players in blood: complement and leukocytes. Biomaterials are known agonists of complement and leukocyte activation, but this is frequently studied only in the context of inflammation. For us, thrombosis is a special case of inflammation. Here we summarize current perspectives on all four of Received 3 September 2003; accepted 19 January 2004 g author. Tel.: +1-416-978-3088; fax: +1-416-978-4317. s:
[email protected] (M.V. Sefton). front matter r 2004 Elsevier Ltd. All rights reserved. materials.2004.01.023 IN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ioma 1. Introduction Biocompatibility is defined as ‘‘the ability of a material to perform with an appropriate host response in a specific application’’. Biocompatibility of blood contacting devices relates mainly to the thrombotic response induced by the materials. Although no material has been found truly biocompatible, many cardiovas- cular devices function with low or acceptable risks of complications [1]. Hemolytic, toxic and immunologic reactions have usually been dealt with earlier in the development of a material to be used for cardiovascular devices and are rarely an issue with their use; an exception may be immunological reactions to tissue engineered constructs. However, thrombotic and throm- boembolic complications, as well as bleeding risks associated with the necessary anticoagulant therapy remain of serious concern with cardiovascular devices. Clinical manifestations of the bioincompatibility of ARTICLE 4. Platelets . . . . . . . . . . . . . . . . . . . . . . . 4.1. Platelet biology . . . . . . . . . . . . . . . . 4.2. Platelets and biomaterials . . . . . . . . . . 5. Leukocytes . . . . . . . . . . . . . . . . . . . . . 5.1. Leukocyte biology . . . . . . . . . . . . . . 5.2. Leukocyte activation and biomaterials . . . . 5.3. Leukocytes, platelets and coagulation . . . . 6. Other important factors . . . . . . . . . . . . . . 6.1. Flow . . . . . . . . . . . . . . . . . . . . . 6.2. Endotoxin . . . . . . . . . . . . . . . . . . 7. Conclusions . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . M.B. Gorbet, M.V. Sefton / B5682 cardiovascular devices are numerous: sudden and complete obstruction of stents within weeks [2]; acute and subacute thrombotic occlusion in medium sized grafts (4–6mm) [3]; embolic complications with artificial hearts [4], catheters [5] and prosthetic valves [6,7]; thrombotic complications during cardiopulmonary by- pass [3] and angioplasty [8]. Larger vascular grafts remain thrombogenic for many years, but fewer thrombotic complications are observed as high flows disperse the clotting factors. However, occasional embolic episodes occur as high flows dislodge the thrombotic deposits. Even if the risk of thrombotic complication appears to be low (varying between 2% and 10% depending on the device), they may have fatal outcomes and the cost associated with the follow-up intervention is not negligible. Furthermore, these thrombotic complications with cardiovascular devices occur despite the use of antiplatelet and anticoagulant therapies reinforcing the inherent thrombogenicity of the materials. Material thrombogenicity is further illustrated by the acute failure of small diameter vascular grafts despite the strong anticoagulant regimen. Many years of intensive research on biomaterials have not yet produced a material, which has proven suitable for this last application. To improve the blood compatibility of cardiovascular devices, surface modifications, such as attachment of antithrombotic agents or the immobilization of poly- ethylene oxide (PEO) have been considered but their success has been limited. Treating surfaces with PEO to reduce protein adsorption and prevent platelet adhesion has remained unproven in terms of thrombogenicity. Different heparin coatings have been developed and some have actually been able to reach the commercial stage in cardiopulmonary bypass [9] and in coronary stents. However, reports on the improvement of in vivo biocompatibility have been mixed [10–16]. Heparinized cardiopulmonary bypass circuits appear to partially reduce the inflammatory response associated with cardiopulmonary bypass [17,18]. But to date, heparin PRESS . . . . . . . . . . . . . . . . . . . . . . . . . . 5689 . . . . . . . . . . . . . . . . . . . . . . . . . . 5689 . . . . . . . . . . . . . . . . . . . . . . . . . . 5690 . . . . . . . . . . . . . . . . . . . . . . . . . . 5692 . . . . . . . . . . . . . . . . . . . . . . . . . . 5692 . . . . . . . . . . . . . . . . . . . . . . . . . . 5693 . . . . . . . . . . . . . . . . . . . . . . . . . . 5694 . . . . . . . . . . . . . . . . . . . . . . . . . . 5694 . . . . . . . . . . . . . . . . . . . . . . . . . . 5694 . . . . . . . . . . . . . . . . . . . . . . . . . . 5695 . . . . . . . . . . . . . . . . . . . . . . . . . . 5696 . . . . . . . . . . . . . . . . . . . . . . . . . . 5697 terials 25 (2004) 5681–5703 coatings have not yet been shown to significantly reduce the number of postoperative complications, improve patient outcome, or reduce hospital stay [16,19,20]. This illustrates another limit of our understanding of blood– material interactions: we do not know how much of an inflammatory and thrombosis response is tolerable or whether any of the changes in normal hemostasis induced by the device result in harmful consequences. Since biomaterial strategies have not resolved the problem, different pharmacological approaches are being investigated. Complement inhibition with the use of sCR1 [21] or anti-C5a antibody [22], serine protease inhibitors such as aprotinin [23,24], platelet receptor antagonists such anti-GPIIb/IIIa [25] and cytokine antibody [26] are newer approaches to reduce thrombo- tic complications with cardiopulmonary bypass. While the results are promising—a partial reduction of the inflammatory or the thrombotic response to cardiovas- cular devices—it is as yet not possible to make conclusions with respect to the overall improvement of device biocompatibility. Unfortunately, most agents (Fig. 2). Thrombin is formed following a cascade of reactions where an inactive factor becomes enzymati- cally active following surface contact or after proteolytic cleavage by other enzymes; the newly activated enzyme then activates another inactive precursor factor. Initia- tion of clotting occurs either by surface-mediated reactions, or through tissue factor (TF) expression by cells. The two systems converge into the common pathway resulting in the formation of fibrin clot upon action of thrombin on fibrinogen. Then, Factor XIII, activated by thrombin, crosslinks and stabilizes the fibrin clot into an insoluble fibrin gel. 2.1. The intrinsic pathway The classic picture (Fig. 2) shows the intrinsic pathway being initiated by contact activation of high molecular weight kininogen (HMWK), prekallikrein and Factor XII: it is commonly said that these molecules require contact with (negatively charged) surfaces for zymogen activation in vitro [27]. Factor XII is activated IN PRESS iomaterials 25 (2004) 5681–5703 5683 affect only one of the players in the blood compatibility response and this is not likely to be sufficient to result in clinical benefits. On the other hand, these inhibitors and antibodies provide valuable information on the mechan- isms involved in the thrombotic complications asso- ciated with cardiovascular devices. Under normal conditions, blood contacts an endothe- lium with anticoagulant and antithrombotic properties. The use of a cardiovascular device represents the introduction of a foreign surface in the circulation, without the properties of the endothelium. Blood– material interactions trigger a complex series of events including protein adsorption, platelet and leukocyte Fig. 1. Overview of blood–material interactions showing the compo- nents relevant to thrombosis. While complement and leukocytes are normally considered under ‘‘inflammation’’ we consider them as participants in thrombosis along with platelets and coagulation. ARTICLE M.B. Gorbet, M.V. Sefton / B activation/adhesion, and the activation of complement and coagulation; they are highly interlinked (Fig. 1). This review outlines the current state of understanding of these phenomena with particular reference to the biomaterial or cardiovascular device as an agonist of these thrombotic reactions. This review is not, however, a catalog of surface modification or biomaterial design strategies that have been used in an attempt to control this aspect of the host response. Rather the focus is on the mechanism of the thrombotic response to a biomaterial, how each component is thought to interact with a biomaterial and how these may interact to produce the observed thrombosis. We are particularly interested in the role of complement and leukocytes, which are not often considered by the biomaterials community to be contributors to thrombosis. Our approach is to treat thrombosis as a special case of inflammation. 2. Coagulation cascade Blood coagulation involves a series of proteolytic reactions resulting in the formation of a fibrin clot by adsorption, FXIIa converts prekallikrein into kallik- rein and with HMWK as a cofactor activates Factor XI to Factor FXIa. Factor XIa activates Factor IX to Factor IXa. Following a cascade of reactions involving among others the intrinsic tenase complex (Factor IXa– Factor VIIIa), prothrombin is cleaved into thrombin. The importance of the intrinsic pathway to normal blood coagulation remains speculative, as the occur- rence of negatively charged surfaces in vivo is limited. Fig. 2. Simplified blood-coagulation cascade as viewed by biomaterial textbooks [117] and older hematology books. The intrinsic system, starting with Factor XII, is shown as a linear cascade of zymogen activation steps in parallel with the extrinsic system that involves TF. TF and platelets are shown as ‘cofactors’ of the process (similar to Ca+2) rather than as central participants. The intrinsic and extrinsic systems converge on the common pathway to produce thrombin and fibrin [used with copyright approval from Elsevier/Academic Press]. Collagen present in the subendothelium after vessel injury could be the surface required for this reaction [28]. Under physiologic conditions, the lack of relevance of the contact activation system is consistent with the fact that deficiencies of the contact proteins, HMWK, prekallikrein and Factor XII, have not been associated with abnormal bleeding [1,27,29]. 2.2. The extrinsic pathway The hematology textbooks center their view of the blood-coagulation cascade, on the TF-dependent path- way (Fig. 3); this perspective has replaced the older perspective that is still used in biomaterials textbooks. The physiological initiator of coagulation is TF, which is expressed on damaged cells at the site of vascular injury. Plasma Factor VII (FVII) binds to TF on the cell ARTICLE IN M.B. Gorbet, M.V. Sefton / Bioma5684 membranes and requires activation to FVIIa to form the extrinsic tenase complex: TF-VIIa complex. FVII is activated by trace amounts of thrombin, FIXa, FXa and TF-VIIa complex, the latter being more likely physio- logically relevant as small amounts of TF–FVIIa are present extravascularly in vivo [30,31]. Picomolar concentrations of FVIIa circulate normally in blood and are also thought to serve as a primer in the initiation of the coagulation cascade by allowing direct formation of TF–FVIIa complex upon TF exposure [32]. TF–FVIIa complex on cell membranes cleaves Factor X into Factor Xa in the presence of calcium. The prothrombi- nase complex can then assemble on the membrane and generate thrombin (common pathway of the coagula- tion cascade). FX is not the only physiological substrate of the TF–FVIIa complex. The TF–FVIIa complex also activates FIX [33]. Fig. 3. Revised (simplified) blood-coagulation cascade as given in a standard hematology textbook [30]. Unlike Fig. 2, there is no intrinsic system or Factor XII. The cascade is not linear with several feedback loops and most importantly it begins with TF and the tenase complex. Platelet involvement is still shown as a ‘cofactor’ and the reactions past thrombin are not shown [used with copyright approval from McGraw- Hill]. The extrinsic and intrinsic pathways are not indepen- dent of each other. When coagulation is initiated by a TF-dependent pathway, the intrinsic tenase remains important, since production of FXa by FIXa–FVIIIa complex has been shown to significantly contribute to thrombin generation [34]. It appears that extrinsic tenase TF–FVIIa is responsible for the onset of coagulation while the intrinsic tenase is the major player in the propagation phase [35]. The activation of FX by FIXa is all the more important since tissue factor pathway inhibitor (TFPI) will reduce the production of FXa by TF-VIIa complex [30,31]. 2.3. Physiologic inhibitors of coagulation Most activated coagulation factors are serine pro- teases. Plasma contains several protease inhibitors, such as a1-protease inhibitor, a2-macroglobulin, heparin cofactor II and antithrombin III, to modulate and inhibit their activity. Antithrombin III is the most important [36]: it neutralizes its target enzymes, FXa and thrombin, by forming a complex with the enzyme in which the enzyme’s active site is blocked. In the absence of heparin, complex formation occurs at a relatively slow rate. However, in the presence of the polysacchar- ide heparin or naturally occurring heparan sulfate (on the endothelium), inhibition rates rise significantly. Antithrombin is also able to inhibit FIXa, FXIa and FXIIa [37]. While TF-VIIa is not efficiently inhibited by antith- rombin, it has its own inhibitor: the TFPI. Plasma TFPI is not the major intravascular pool of TFPI. A larger pool exists on the luminal surface of the vascular endothelium, which is released by a bolus injection of heparin. Platelets also carry 10% of the total TFPI in blood and release their TFPI following stimulation by thrombin and other agonists [38]. Activated monocytes may also release TFPI [39]. TFPI has two inhibitory sites, one for FXa and one for TF-VIIa complex. Inactivation of FXa through binding to TFPI in solution is required for inhibition of the complex TF-VIIa. The initial binding of FXa to TFPI is relatively slow [40] and may not be able to prevent thrombosis when TF–FVIIa complex are being formed at a high rate [31]. Leukocyte elastase also cleaves TFPI and impairs its ability to inhibit both FXa and TF-VIIa complex. Upon exposure of a complex FXa/TFPI/TF- VIIa to elastase, FXa and TF-VIIa activities are restored [41]. The endothelium also participates in the regulation of thrombosis via thrombomodulin. Thrombin binds to PRESS terials 25 (2004) 5681–5703 thrombomodulin on endothelial cells and this complex activates protein C [42]. Protein C is a vitamin K- dependent protein and activated protein C inactivates FVa and FVIIIa. Protein S, also a vitamin K-dependent protein, is a necessary cofactor for activated protein C. IN ioma More than half of the protein S in plasma is bound to the C4b-binding protein (a regulatory protein of the complement system) and is not functionally active. 2.4. Biomaterials and coagulation pathways While under physiological conditions the role of FXII activation is questionable, in the presence of a cardio- vascular device activation of FXII may occur. The artificial surface, with its adsorbed protein layer, then represents the required (‘‘negatively charged’’) surface. Protein adsorption is the first event in blood–material interactions, and adsorption of the contact phase proteins may result in activation of the intrinsic cascade. Earlier studies had focused on protein adsorption on glass or biomaterial surfaces with isolated protein solutions or diluted plasma, and showed how fibrinogen is replaced over time by HMWK (the Vroman effect), suggesting a possible role of the intrinsic pathway in material thrombosis [43]. Recent studies using whole blood or plasma have provided new insights on the adsorption and activation of contact phase proteins. FXII adsorption has been observed in moderate amounts on materials used in vascular grafts [44] and hemodialysers [45]; however, it was not found in its activated form [46]. Although HMWK and prekallikrein may be adsorbed on the material surface, the lack of FXIIa on the material surface will stall the initiation of coagulation through the intrinsic pathway. In some instances, in vitro activation of FXII and kallikrein has been reported with biomaterials [47–49]; however, no test was performed to determine if such activation resulted in any significant activation of coagulation. Other studies have actually shown that only minute amounts of thrombin or thrombin–antithrombin III complex (TAT) are generated when biomaterials are incubated with undiluted plasma alone [50–52]. Further- more, higher levels of adsorbed kallikrein and Factor XII on biomaterials do not correlate with TAT formation [50], suggesting that the contact phase proteins, by themselves, play little role in the activation of coagulation. In fact, a study by Hong et al. [51] suggested that the presence of leukocytes is required for activation of the coagulation cascade; the requirement of a TF-dependent pathway of initiation of coagulation may thus also apply to biomaterials. In vivo, a small increase of FXIIa is observed during cardiopulmonary bypass [53,54], but it appears to be in response to the surgical intervention and the establishment of extra- corporeal circulation (i.e., exposure to the biomaterials of the circuit) does not further increase FXIIa levels [54]. ARTICLE M.B. Gorbet, M.V. Sefton / B Furthermore, no significant correlation has been ob- served between FXIIa and thrombin generation [54]. In vivo results with hemodialysis also failed to show any significant increase of FXIIa [55]. That a patient with a severe FXII deficiency showed levels of thrombin generation during cardiopulmonary bypass similar to normal patients [56] casts further doubt on the role of Factor XII in the initiation of coagulation with biomaterials. Taken together, these studies do not support the view that activation of the contact phase proteins is important in the activation of coagulation by biomaterials. While the TF pathway of blood coagulation has become the focus in hematology and has led to the revised version of the coagulation cascade (compare Figs. 2 and 3), the biomaterials community has been slow to recognize its importance. The textbooks refer to the older model of coagulation with the separated intrinsic and extrinsic pathways with the thought that the extrinsic pathway is not directly related to blood– material interactions [1,57–59]. However, blood contact with a material represents a potential stimulus to induce TF expression by monocytes, resulting in blood coagulation through the extrinsic system. Indeed, monocyte TF expression has been observed in vivo during or after cardiopulmonary bypass [60–62]. Further details of TF expression in the presence of biomaterials are given elsewhere [63,64] and summarized below in the section on leukocytes. The role of leukocytes (most likely due to TF on monocytes) in activation of the coagulation by biomaterials was highlighted in the study by Hong et al. [51] that was referred to earlier. Thrombin–antithrombin formation (TAT) on PVC was found to be negligible in both plasma and platelet-rich plasma, while significant levels of TAT were observed in whole blood; i.e., only in the presence of leukocytes was there significant thrombin formation. More research is needed, however, to define the relative importance of the extrinsic and intrinsic pathways of coagulation in the overall picture of thrombosis on biomaterials. The time scales associated with initiation of the coagulation cascade by contact phase activation and with TF expression will have an impact on their relative importance. As it is part of the protein adsorption ‘‘reaction’’, contact phase activation occurs during the first few seconds/minutes of blood contact with a material. On the other hand, to be expressed on monocytes, TF requires synthesis and thus a minimum of 60min (in a normal patient) would elapse before this pathway could significantly contribute to thrombin generation. Thrombin generation by FXIIa on materials is also dependent on flow [65], because of the effects of flow on mass transfer as well as direct effects on platelet/leukocyte phenotype. One manifestation of this is the relevant time scales of the coagulation PRESS terials 25 (2004) 5681–5703 5685 reactions will vary with the flow situation. Thus, the relative roles of the intrinsic and extrinsic pathways in thrombin generation will likely depend on both the flow situation and the relevant time frame. This adds another level of complexity to the understanding/ complement activation. IN ioma characterization of thrombin generation with cardio- vascular devices. 2.5. Anticoagulants To prevent the formation of thrombi during cardio- pulmonary bypass, hemodialysis and angioplasty, antic- oagulants are routinely administered. Heparin is the most common [37]. Heparin binds to antithrombin III (ATIII) via a unique pentasaccharide sequence, and causes a conformational change in the reactive center of ATIII, thereby accelerating the rate of ATIII- mediated inactivation of several clotting enzymes (thrombin, FXIa, FXa and FIXa). Heparin promotes the formation of the complex thrombin-ATIII (TAT) by binding to both proteins. On the other hand, to inactivate FXa, heparin needs to bind only to ATIII. ATIII forms a 1:1 irreversible complex with coagulation enzymes, the heparin then dissociates and can be reused. A limitation of heparin is that it is unable to inactivate thrombin bound to fibrin or to surfaces, or to inhibit FXa within the prothrombinase complex [66]. Heparin can also cause thrombocytopenia. Heparin binds to platelet factor 4 (PF4) and in some patients, antibodies will develop against this heparin–PF4 complex. The antibody–PF4–heparin complex then binds to platelets and induces platelet activation, aggregation and acti- vates the blood-coagulation pathways, resulting in both a loss of circulating platelets and a thrombotic state [67]. Following implantation of cardiovascular devices, such as vascular grafts and artificial valves, antic- oagulants such as warfarin is used. It can be taken orally and it interferes with the vitamin K cycle, thus impairing the biological function of vitamin K-depen- dent coagulation proteins (prothrombin, FVII, FIX and FX) [37]. For stents, anticoagulants are not needed except during placement. Aspirin and Plavix (clopido- grel, antiplatelet aggregation) are used after placement (aspirin forever; Plavix for 1 month or 1 year) to control thrombosis, while heparin and a GPIIb/IIIa antagonist (e.g., ReoPro) is only needed perioperatively. Other anticoagulants of interest for use of cardiovas- cular devices [66] are the tick anticoagulant peptide (TAP) and antistatin which binds to FXa even within the prothrombinase complex; hirudin, a leech-derived protein, a potent thrombin inhibitor; and D-Phe-Pro- ArgCH2Cl (or PPACK), a peptide that directly inacti- vates thrombin by interacting with the active site of thrombin. Both PPACK and hirudin are able to inactivate fibrin-bound and surface-bound thrombin ARTICLE M.B. Gorbet, M.V. Sefton / B5686 [66,68] which is a significant advantage. Ximelagatran is also a new oral direct thrombin inhibitor and clinical trials have been very successful: similar or superior efficacy relative to warfarin in some scenarios with reduced bleeding, obviating the need for monitoring. It 3.1. Classical pathway The classical pathway is normally triggered by antigen–antibody complexes that bind the C1 complex (C1q, C1r, C1s) through the C1q component. This activates C1s, which is then able to cleave the C4 complement protein into C4a and C4b. C4b attaches to its target surface via its exposed metastable thioester binding site. It is important to note that C4b does not bind efficiently to membrane surfaces and the fluid has many advantages over warfarin and will likely soon replace it [69]. When studying blood–material interactions in vitro, heparin is usually the anticoagulant of choice as it is the most widely used with cardiovascular devices. However, heparin also possesses some anticomplement activity [70]. PPACK and hirudin, which appear to be more specific than heparin, may then be used especially when the mechanisms of cell activation are studied. However, the associated high cost restricts their use. 3. Complement The complement system plays an important role in the body’s defense mechanisms against infection and ‘‘non-self’’ elements. The complement system consists of more than 20 plasma proteins that function either as enzymes or as binding proteins. Complement activation is initiated via the classical or alternative pathways and the terminal pathway is common to both (Fig. 4). Both pathways contain an initial enzyme that catalyses the formation of the C3 convertase, which in turn generates the C5 convertase allowing the assembly of the terminal complement complex (TCC). Various complement products (C3b, C4b and iC3b) bind to particles, surfaces, bacteria and immune com- plexes in a process called opsonization [71], which facilitates their uptake by inflammatory cells. Activation of complement results in cell lysis when the terminal attack complex is inserted into the cell membrane. Complement activation also releases C3a, C4a and C5a, which are anaphylatoxins. These peptides are humoral messengers that bind to specific receptors on neutro- phils, monocytes, macrophages, mast cells and smooth muscle cells. They induce a variety of cellular responses such as chemotaxis, vasodilatation, cell activation and cell adhesion [72]. At high enough concentration, they are responsible for the many systemic effects of PRESS terials 25 (2004) 5681–5703 phase C4b is rapidly inactivated by the loss of its binding site. C2 binds to the attached C4b and is cleaved by C1s, releasing C2a. The classical C3 convertase, C4bC2b, is thus formed and can cleave C3 into C3a (anaphylatoxin) and C3b. The combination of C4bC2b IN ioma ARTICLE M.B. Gorbet, M.V. Sefton / B and C3b becomes the C5 convertase, which cleaves C5 into C5a (anaphylatoxin) and C5b. C5b is the first component of the terminal complex and has high affinity for C6. C5bC6 then binds C7, C8 and up to 12 molecules of C9 and this forms the TCC C5b-9. If C5b is attached to a biological surface, the TCC (also called the membrane attack complex, mC5b-9) inserts itself into the lipid layers resulting in cell damage and/or lysis. In the absence of a biological membrane, the complex binds to S protein (also known as vitronectin) to create SC5b-9 in the fluid phase. 3.2. Alternative pathway The activation of the alternative pathway does not require antibody or immune complexes and is activated by any foreign surfaces, such as fungal, bacterial polysaccharides, lipopolysaccharides (LPSs), particles and biomaterial surfaces. Complement activation via the alternative pathway occurs spontaneously at a low rate. Fig. 4. Pathways of (a) complement activation [256], copyright JB Lippinco alternative pathways are shown leading to the production of the C3 convertas loops. In (b), various natural and synthetic inhibitors are shown inhibiting d inhibitor is shown as being effective at a level prior to the formation of the PRESS terials 25 (2004) 5681–5703 5687 Spontaneous hydrolysis of the internal thioester group of C3 occurs in the fluid phase, generating C3 �H2O. This hydrolyzed C3 can bind and activate Factor B and cleave another C3 molecule into C3a and C3b. The alternative C3 convertase, C3bBb, is formed. In the absence of a surface to support binding of C3b, little complement activation occurs. In the presence of a surface, covalent binding of C3b to hydroxyl or amine groups on the surface may occur via the carbonyl group in the C3b thioester binding site. Attachment of C3b to a surface favors binding of Factor B and Factor D to C3b. Factor D cleaves Factor B into Ba and Bb, and the alternative C3 convertase, C3bBb, is formed again. This attached C3 convertase is able to generate more C3b, resulting in a positive amplification loop. Properdin acts to stabilize the C3 convertase. The clustering of C3b on the surface allows the formation of the alternative C5 convertase, C3bBbC3b, and C5 can be cleaved. The assembly of the TCC follows as for the classical pathway. tt and (b) inhibition [257], copyright Elsevier. In (a), the classical and e and C5 convertase complexes, both of which are part of amplification ownstream portions (C3 and C5) of the complement cascades; only C1 C3 or C5 convertases. such as hydroxyl and amino groups: these groups will allow covalent binding of C3b and promote formation of the C3 and C5 convertase [2,73]. However, even in the absence of these activating groups on the surface, some biomaterials, such as polyacrylonitrile, are able to activate complement, suggesting that the mechanisms of material-induced complement activation due to nucleophilic groups is not the whole story. A newer hypothesis places emphasis on interaction of Factor H with the surface: an activating material is then defined by its capacity to bind Factor B rather than Factor H [73]. As noted earlier, binding of Factor H would lead to C3b inactivation by Factor I and thus terminate the propagation of the complement cascade. IN ioma 3.3. Regulatory molecules of complement activation The various molecules involved in the regulation of complement activation are illustrated in Fig. 4b. The classical pathway is regulated by two specific mechan- isms [29]: C1 inhibitor, a plasma protein, binding C1s or C1q and inhibiting the enzymatic activity of the C1 complex; and the C4b-binding protein inhibiting C4b bound to a membrane or a surface. Factor I is a proteolytic enzyme that binds C4b and C3b, generating iC4b and iC3b, which are further degraded into smaller fragments. C4b-binding protein is a cofactor of Factor I and augments the degradation of C4b by Factor I. Factor H is a cofactor of Factor I for the degradation of C3b. Factor H is also able to displace Bb in the C3 convertase to promote C3b inactivation by Factor I [75]. On cell surfaces, the C3 and C5 convertase activity are regulated by three integral membrane proteins [75]. Decay-accelerating factor (DAF), found on all periph- eral blood cells, destabilizes the C3 convertase by promoting the release of factor Bb. The membrane cofactor protein (MCP), expressed on leukocytes and platelets, favors the dissociation of Factor B and promotes C3b association with Factor I. The comple- ment receptor type 1 (CR1), present on all blood cells except platelets, acts like Factor H and displaces Bb from the C3 convertase and facilitates inactivation by Factor I. To prevent lysis of ‘‘bystander’’ blood cells, mem- brane proteins called homologous restriction factors limit the ability of the terminal complex to properly assemble on autologous cells [75]. Two proteins have been characterized: CD59 and the C8-binding protein (also called MAC inhibiting protein). CD59, found not only on blood cells but on many cells such as hepatocytes and epithelial cells, binds to C8 and C9 and inhibits C9 polymerization. Little is known about the C8-binding protein, which is believed to bind C8 and C9. 3.4. Interactions of complement and coagulation cascade Although the coagulation and complement cascades are discussed as separate entities, the two cascades appear to interact significantly to modulate each other’s activity [29,75]. Factor XIIa and kallikrein are known to cleave C1s [29] and thus have the capacity to trigger classical complement activation. Thrombin activates C3, C5, C6 and Factor B; kallikrein cleaves C5 and factor B; and Factor XIIa also cleaves C3. The ARTICLE M.B. Gorbet, M.V. Sefton / B5688 activity of thrombin on C3 and C5 may actually explain the higher background levels of C3a and C5a in serum versus plasma. Table 1 outlines the various interactions between complement and coagulation factors. 3.5. Complement activation and biomaterials Complement activation is generally treated as if it is part of the inflammatory response induced by biomater- ials. For example, complement activation is known to occur during cardiopulmonary bypass and hemodialysis [75–78], and with catheters and prosthetic vascular grafts [79,80]. It is recognized that, both in the short and long term, complement activation plays a role in the leukocyte related clinical sequelae associated with the use of cardiovascular devices such as leukopenia, hypotension and pulmonary injury [81,82]. The throm- botic consequences of leukocyte activation are discussed below. The presence of a biomaterial is believed convention- ally to activate complement via the alternative pathway. Biomaterials are usually classified as ‘‘activating’’ or ‘‘non-activating’’ surfaces [74]. On a non-activating surface, negatively charged groups such as carboxyl and sulfate, sialic acid and bound heparin appear to promote high-affinity association between bound C3b and Factor H. On the other hand, an activating surface is usually characterized by the presence of nucleophiles PRESS Table 1 Interactions between complement and coagulation systems [29,75] Protein Type of interaction Thrombin Proteolysis of C3, C5, C6 and factor B Factor XIIa Proteolysis of C1r, C1s and C3 Kallikrein Proteolysis of C1, C5 and Factor B Antithrombin III Protect RBC from lysis by mC5b-9 Bb Proteolysis of prothrombin C3bBb Proteolysis of prothrombin C1 inhibitor Inactivates FXIIa and kallikrein S protein (vitronectin) Stabilizes plasminogen activator inhibitor 1 C4b-binding protein Binds to the vitamin K-dependent protein S terials 25 (2004) 5681–5703 Some activating materials generate high levels of both C3b and C5b-9, while others will generate high C3b level but little C5b-9. Why the efficiency of C5 convertase formation relative to that of C3 formation differs from one activating surface to another is not well understood. host will be able to differentiate/discriminate between moderate and high complement activating surfaces. Platelets respond to minimal stimulation and become activated when they contact any thrombogenic surface such as injured endothelium, subendothelium and artificial surfaces. Platelet activation is initiated by the interaction of an extracellular stimulus with the platelet surface. This interaction involves the coupling of the agonist to specific receptors on the platelet plasma membrane [91]. Plasma proteins such as thrombin and fibrinogen; vascular wall products such as collagen; and molecules derived from inflammatory cells (i.e., leuko- cytes) or platelets, such as platelet activating factor (PAF) or cathepsin G, are all potent platelet activators. A list of known platelet receptors and their specific agonist/ligand is presented in Table 2. Activation results in at least five physiologic responses [92]. (1) A platelet release reaction in which biologically active compounds stored in intracellular platelet gran- ules are secreted into the microenvironment, such as platelet factor 4, thrombospondin, b-thromboglobulin, ADP and serotonin. (2) P-selectin (previously referred to as GMP-140 or PADGEM) is released and expressed on the platelet membrane after a granule secretion. IN PRESS Table 2 Platelet receptors [91,102] Receptor Ligand/agonist (a) Receptors leading to platelet activation Thrombin receptor Thrombin iomaterials 25 (2004) 5681–5703 5689 The hypothesis that material-induced complement activation occurs exclusively via the alternative pathway has also been challenged. Reports of complement activation via the classical pathway during cardiopul- monary bypass [18,83] and a delay in complement activation observed with C4-deficient patients under- going hemodialysis [84] suggest that classical activation plays a role in material-induced activation. The presence of immune complexes may allow for a rapid onset of complement activation, and then subsequently the alternative pathway becomes activated. In vitro studies have also demonstrated activation of the classical pathway by some biomaterials [85–87]. C1 inhibitors such as pentamidine and benzamidine were effective in lowering platelet adhesion and activation by polystyrene beads and polyethylene tubes while sCR1, an inhibitor of both pathways at the level of C3, had no effect [88,89]. Pentamidine was also effective in a canine chronic shunt in eliminating the thrombocytopenia seen with a platelet activating material (a polyvinyl alcohol (PVA) hydrogel) [90]. While some of the consequences of complement activation are well understood, more work is needed to fully understand how a material activates complement. Questions to be resolved include selecting inhibitors that block both pathways and at an early enough stage to inhibit the local (rather than systemic) effects of complement activation. Similarly, controlling the differ- ential adsorption of C3b, Factor D and Factor H may be more important than preparing low adsorption, so-called ‘‘non-fouling’’ surfaces. The latter may lower the adsorption of all proteins but it may be more important to alter the composition of the protein adsorbate. 4. Platelets 4.1. Platelet biology The platelet’s main role in hemostasis is to preserve the integrity of the vascular wall through formation of a platelet plug. Platelets are anuclear, disc-shaped cells However, even low amounts of C5b-9 are able to activate leukocytes [74] and thus a low terminal complement activating material may still induce a significant inflammatory response. The question remains as to which are the appropriate levels of complement activation that the host can accept without deleterious effect. We also have to determine if, in the long term, the ARTICLE M.B. Gorbet, M.V. Sefton / B with a diameter of 3–4 mm. They are derived from megakaryocytes in the bone marrow and circulate at an average concentration of 200� 106 cells/ml, with in- dividual platelet concentrations ranging from 150 to 400� 106 platelets/ml. Thromboxane A2 receptor TxA2, PGH2, PGG2 V1A receptor Vasopressin PAF receptor Platelet activating factor 5HT2 receptor Serotonin or 5-OH tryptamine a2 receptor Epinephrine ADP receptor ADP, ATP PGE2 receptor PGE2 C1q receptor C1q (b) Receptors leading to platelet inhibition A2 receptor Adenosine PGI2 receptor PGI2, PGE1 PGD2 receptor PGD2 (c) Platelet adhesion receptors (binding may also result in platelet activation) GPIa/IIa or VLA-2 Collagen GPIb/IX or GPIb Von Willebrand factor (vWF), thrombin GPIc/IIa or VLA-5 Fibronectin GPIc0/IIa or VLA-6 Laminin GPIIb/IIIa Collagen, fibrinogen, fibronectin, vitronectin, vWF GPIV or GPIIIb Collagen, thrombospondin, GPVI Collagen Vitronectin receptor Vitronectin, thrombospondin PECAM-1 Heparin Fcg-RII Immune complexes ICAM-2 LFA-1 P-selectin Sialyl-Lex Leukosialin, sialophorin ICAM-1 GP: glycoprotein; VLA: very late antigen. IN ioma P-selectin is a cell-surface glycoprotein belonging to the selectin family and plays an important role in mediating adhesion of activated platelets to neutrophils, mono- cytes and a subset of lymphocytes [93,94]. (3) The platelet eicosanoid pathway is initiated, resulting in the liberation of arachidonic acid from platelet phospholi- pids and in the synthesis and release of prostaglandins and thromboxane B2. (4) Platelet activation is char- acterized by a drastic shape change, which promotes platelet–platelet contact and adhesion. The rearrange- ment of the platelet membrane during activation also promotes the association of the tenase and prothrombi- nase complexes on its phospholipids. (5) Platelet activation results in the formation of platelet micro- particles (PMPs), which are particularly rich in factor Va, platelet factor 3 and phospholipid-like procoagulant activity (phosphatidylserine) [95,96]. PMPs are formed from the surface membrane through exocytotic budding. Their physiologic role remains unclear but in vitro results have shown that they can bind and adhere to fibrinogen and fibrin, and coaggregate with platelets [97,98]. The procoagulant activity of PMPs, generated both in vitro and in vivo, has also been demonstrated [99–101]. Among the different platelet adhesion receptors (Table 2), GPIb and GPIIb/IIIa have the highest density on platelets. GPIb (CD42) is a leucine-rich glycoprotein receptor and approximately 25,000 receptors are present on the platelet surface [92,102]. It is complexed one to one with GPIX but the function of the latter remains unknown. GPIb is a long molecule, making it more susceptible to conformational change upon shear stress. GPIb mediates platelet interaction with von Willebrand factor (vWF). It will not bind plasma vWF unless the antibiotic ristocetin or the snake venom botrocetin is present. On the other hand, GPIb will bind to adsorbed or immobilized vWF on a surface. Shear stress is an important factor in platelet adhesion to vWF as it induces the required conformational change of vWF to bind GPIb. Thrombin also binds to GPIb but the functional significance of this binding has not been elucidated. GPIIb/IIIa (CD41/CD61) is an integrin receptor and is constitutively expressed on platelets. GPIIb/IIIa is the dominant platelet receptor with 40–80,000 receptors present on the surface of a resting platelet. Another 20–40,000 are present inside the platelets, in a granule membranes and in the membranes lining the open canalicular system [102]. They are translocated to the platelet membrane during the release reaction. On resting platelets, GPIIb/IIIa is in an inactive form and ARTICLE M.B. Gorbet, M.V. Sefton / B5690 has a low-affinity binding site for adsorbed fibrinogen [103]. Upon platelet activation, a conformational change occurs leading to the exposure of the high-affinity binding site for soluble fibrinogen. Binding of fibrinogen to GPIIb/IIIa leads to platelet aggregation as well as formation of platelet–leukocyte aggregates, via cross- linking of GPIIb/IIIa on two different platelets by fibrinogen and crosslinking between GPIIb/IIIa and Mac-1 (on leukocyte) by fibrinogen. Other adhesive glycoproteins containing RGD sequences can also bind to activated GPIIb/IIIa: vWF, thrombospondin, fibro- nectin and vitronectin. Since GPIIb/IIIa mediates platelet aggregation, its inhibition has generated much interest in the control of prothrombotic states [104,105]. An antibody against GPII/IIIa (7E3 also called Reopro or Abciximab) has been developed and has entered clinical trials with angioplasty [106], myocardial infarc- tion [107] and unstable angina [107,108]. All clinical trials have shown a significant improvement of long- term survival [109,110]. However, higher complication rates such as bleeding and thrombocytopenia have also been observed [111,112]. New clinical trials are under- way to determine appropriate regimens [110]. 4.2. Platelets and biomaterials Platelet activation (platelet release, PMP formation, P-selectin expression, aggregation) and adhesion is known to occur [1,58] during cardiopulmonary bypass, hemodialysis, as well as with vascular grafts and catheters. The thrombotic complications associated with cardiovascular devices are linked clearly to their ability to activate platelets. Adherent platelets [113] and circulating PMPs generated by material contact [100,101] have been shown to be procoagulant in nature. Association between platelets and leukocytes via P- selectin also occurs in the presence of cardiovascular devices [8] and such associations have become a relatively new parameter to study biocompatibility. However, the implications of this association are mostly unknown: they may directly or indirectly contribute to thrombin generation (via monocyte TF) or participate in the removal of platelets from the circulation since several platelets can be bound to each neutrophil or monocyte. While it is intuitive to suggest that a non-thrombo- genic surface should not support platelet adhesion it has not, unfortunately, been that simple. It has been found that platelet contacts with some biomaterial surfaces results in activation leading to high consumption (removal from circulation) characterized by high platelet turnover rather than adherence and by the formation of microemboli rather than occlusive thrombi. Following contact with the layer of adsorbed proteins on the artificial surface, platelets will either adhere or ‘‘bounce off’’ [114], most likely depending on their state of PRESS terials 25 (2004) 5681–5703 activation and the ligands present at the interface [115]. Platelet adhesion on surfaces is mediated by GPIIb/IIIa and fibrinogen and interaction with GPIb/IIa and vWF can also occur [116–119]. However, the absence of significant platelet adhesion does not preclude platelet IN ioma activation as shown by the generation of PMPs with PVA hydrogel both in vitro [120] and ex vivo [90]. Indeed, animal studies have shown that, despite the absence of platelet adhesion, blood contact with some hydrogel surfaces [121,122], like contact with NHLBI reference materials and Silastics [123] appears to activate platelets, resulting in their removal from the circulation. Furthermore, Hanson et al. [122] demon- strated a direct linear relationship between the water content of hydrogels and the rate of platelet consump- tion in a baboon AV shunt model. They also noted that platelet consumption was reduced by the antiplatelet agent, dipyridamole [124], similar to its effectiveness in normalizing platelet survival in-patients with artificial heart valves [125]. While there is agreement that the rapidly adsorbed proteins, especially fibrinogen, play a critical role in platelet adhesion, it is not clear, for example, how effective adsorbed fibrinogen is as a platelet agonist in vivo and what other mechanisms are involved in supporting or initiating platelet adhesion. While ad- sorbed fibrinogen is likely the critical ligand for adhesion, what activates the platelet to adhere in the first place? Further it has not been clear that inhibition of platelet adhesion will inhibit the generation of PMPs or reduce platelet consumption. The strategies available to minimize platelet adhesion (e.g., polyethylene glycol immobilization) have not been sufficient to prevent platelet consumption [126]. The mechanism of material- induced platelet activation is often presumed to be via the generation of thrombin due to activation of the intrinsic coagulation cascade or the release of ADP from damaged red blood cells or platelets. Even in the presence of heparin, small levels of thrombin generation are generated and may activate platelets. However, the inability of thrombin and kallikrein inhibitors to reduce platelet activation suggests that platelet activation is at least in part mediated by other agonists [127]. For example, a correlation between complement activation and thrombocytopenia has been noted during dialysis [128,129]. Complement activation can lead to platelet activation in many ways. Platelets possess a receptor for C1q that has been shown to induce GPIIb/IIIa activation, P-selectin expression and procoagulant activity [130]. It is currently not known how classical complement activation leads to activated platelets, but in vitro results support a role for C1 [89]. Insertion of C5b-9 in platelets has also been associated with increased platelet procoa- gulant activity [95]. In vitro studies using human cells have been conducted to probe the effectiveness of ARTICLE M.B. Gorbet, M.V. Sefton / B various agents to inhibit platelet activation [88] using a microsphere based immunoassay. Platelet adhesion to polystyrene microspheres was found to be unaffected by the complement inhibitor, sCR1 which is otherwise capable of inhibiting material-induced SC5b-9 produc- tion. In contrast, classical pathway complement inhibi- tors, pentamidine, benzamidine, pyridoxal-5-phosphate (P5P) and cysteine were able to inhibit platelet adhesion to the polystyrene surface. These agents also inhibited platelet loss and microparticle levels in whole blood after contact within polyethylene tubing. Benzamidine and a derivative, pentamidine, are able to competitively inhibit C1s enzyme (ionic interaction with active site), although pentamidine has an IC50 10 times lower [131–133]. The antiplatelet effects of pentamidine have been documented but the mechanism of action remains unresolved. It has been reported that pentamidine has no disruptive effect on GPIIb/IIIa receptors, intracellular cAMP levels, calcium ion influ- xes or intracellular pH changes [88,134,135]. P5P, a major coenzyme from Vitamin B6 is also a known inhibitor of C1 fixation [136,137]. Cysteine is known to inhibit both C1s and the alternative complement path- way [138]. Since sCR1 (in vitro) was not able to block platelet adhesion and activation, it is our hypothesis that C1 is playing a role in material-induced platelet activation and that these agents are effective because of their ability to inhibit C1s. These agents also effectively inhibit other serine proteases such as throm- bin, trypsin and plasmin [132] and thus unequivocal delineation of the mechanism is not yet possible. However, during in vitro blood–material contact, inhibiting complement activation has led to conflicting results on the role of the terminal complement pathway in material-induced platelet activation. Monoclonal antibodies to C5 and C8 inhibited platelet activation during simulated extracorporeal circulation (SECC) [139,140]. On the other hand, sCR1 had no effect on platelets in our microsphere assay [88,141] and in a different extracorporeal circulation model [142]. The difference in experimental conditions such as higher flow rate, blood dilution and the presence of mannitol (a hydroxyl scavenger) between the studies may account for the difference in the apparent efficacy of terminal pathway inhibition. There are likely multiple pathways whereby platelets are activated, some more relevant during conditions of high complement and leukocyte activation and some more relevant where these path- ways are less well developed. Further research is needed to fully understand which complement protein triggers material-induced platelet activation. In our arterio-venous canine shunt model with a platelet consumptive PVA hydrogel tubing segment, systemic low molecular weight heparin did not reduce material-induced platelet damage: loss of platelets, microparticle generation or reduced lifespan, indicating PRESS terials 25 (2004) 5681–5703 5691 that thrombin production did not appear to play a role. On the other hand, pentamidine (12mg/kg, daily, IM) dramatically inhibited thrombocytopenia during the connection of PVA hydrogel test segments (Gemmell, 2001, pers. comm.). This supports our hypothesis that such inhibitors are useful for blocking biomaterial- associated platelet activation. We have also used ReoProTM (up to 0.8mg/kg, GPIIb/IIIa antagonist) and not observed any effect on thrombocytopenia caused by PVA. The ReoProTM dosage was sufficient to inhibit agonist induced platelet aggregation. This preliminary finding could suggest that platelet adhesion (and vesiculation)—both blocked by IIb/IIIa inhibi- tors—are not responsible for premature platelet con- sumption and clearance, at least on these smooth surfaces that have few platelet deposits in the absence of any therapeutic agents. macrophages. It is important to note that when neutrophils and monocytes are recruited in tissues during inflammation, their half-life increases to several days. Monocytes and neutrophils possess receptors for different complement products and other pro-inflam- matory mediators such as PAF and cytokines. Platelet release, such as b-thromboglobulin and PDGF, has also been reported to activate neutrophils in vitro [145–147]. Other neutrophil and monocyte activating stimuli include bacteria and their products and cell adhesion. A list of the most relevant receptors involved in the inflammatory response is presented in Table 3. As for platelet activation, leukocyte activation results ARTICLE IN PRESS 5] M.B. Gorbet, M.V. Sefton / Biomaterials 25 (2004) 5681–57035692 Table 3 Leukocyte receptors in acute/immediate inflammatory response [254,25 Receptor Ligand (a) Complement receptors C1q R C1q, MBP CR1 C3b>C4b>iC3b CR3 or CD11b iC3b, fibrinogen, FX, ICAM-1 CR4 or CD11c iC3b, fibrinogen C3a R C3a C5a R C5a 5. Leukocytes 5.1. Leukocyte biology Circulating leukocytes comprise neutrophils, mono- cytes, lymphocytes, basophils and eosinophils. Only neutrophils and monocytes in blood, but not after they emigrate into tissues, will be addressed in this review as they are the major players in the inflammatory response with cardiovascular devices. Neutrophils are the most abundant white blood cells, representing 40–60% of the leukocyte population (3–5� 106 neutrophils/ml), while monocytes represent 5% with a concentration of 0.2–1� 106 monocytes/ml. Under normal circumstances, neutrophils have a very short half-life in blood (8–20 h) but after an inflamma- tory stimulus, such as LPS or cytokines, their lifespan can increase up to three-fold [143] and their role may be even more active. Monocytes enter the circulation for a short period (36–104 h) and they migrate into tissues where maturation and differentiation occur and they become macrophages [144]. They may also be deposited on injured blood vessels and later differentiate into (b) Other receptors (R) TNFa , IL-1 IL-8 PAF, LTB4 GM-CSF, G-CSF, IFNg in several physiological responses. (1) Upon activation, changes in expression of membrane receptors occur on neutrophils and monocytes: upregulation of CD11b by translocation from intracellular granules [148], shedding of l-selectin by shedding [148], synthesis and expression of TF [149]. TF expression by leukocytes is the subject of another review [63]. (2) Leukocyte activation results in release of inflammatory mediators. Neutrophils contain three types of granules (gelatinase, specific and azurophil granules) and their contents may be released upon activation: among others, elastase, cathepsin G and lactoferrin are important inflammatory mediators. Cytokines such as IL-1, IL-6, IL-8, TNFa, G-CSF and GM-CSF are also released. Arachidonic acid metabo- lites, such as leukotriene B4 and PAF, are produced and released by activated neutrophils. The released inflam- matory mediators have various properties: they may be chemoattractant for leukocytes, promote adherence to endothelial cells, and further activate platelets or leukocytes. (3) Activation may also result in the onset of the oxidative burst whereby neutrophils and monocytes release oxidants, such as O2 � and H2O2. These products damage tissues and activate cells [150]. Function Enhance phagocytosis, respiratory burst Immune adherence, phagocytosis Phagocytosis, respiratory burst, adhesion Adhesion, phagocytosis Chemotaxis, degranulation, respiratory burst Chemotaxis, degranulation, respiratory burst Degranulation, respiratory burst Chemotaxis, degranulation, respiratory burst Strong activation Weak activation, priming IN ioma (4) Activated neutrophils and monocytes also have an increased adhesive capacity on endothelium and other surfaces [151]. Leukocyte adhesion to the endothelium is an impor- tant means by which neutrophils and monocytes participate in the inflammatory response. Adherent leukocytes have been shown to be more activated than their counterpart in the bulk [152,153] but the level of activation of adherent leukocytes depends on the surface ARTICLE Table 4 Leukocyte adhesion receptors Receptor Ligand (a) Selectin family l-selectin Mucin-like ligand, lymph node addressin (b) Integrin family CD11a/CD18 (or LFA-1) ICAM-1, ICAM-2, ICAM-3 CD11b/CD18 (CR3 or Mac-1) iC3b, fibrinogen, FX, ICAM-1 CD11c/CD18 (CR4 or p150,95) iC3b, fibrinogen VLA-1 Laminin, collagen VLA-2 Collagen VLA-3 Fibronectin, laminin, collagen VLA-4 VCAM-1 VLA-5 Fibronectin (c) Immunoglobulin family PECAM-1 PECAM-1, heparin Fcg-RI, Fcg-RII, Fcg-RIII Immune complexes VLA: very late antigen. M.B. Gorbet, M.V. Sefton / B [154]. Leukocyte adhesion molecules are divided into three main families [155]: the selectins, the integrins and the immunoglobulin superfamily (Table 4). The mechan- ism of leukocyte adhesion to endothelial cells, a three- step mechanism, has been well characterized [156]. Step 1: l-selectin is involved in the initial rolling of leukocytes on endothelium. Step 2: The rolling stage enables leukocytes to slow their movement and sample the local environment, and they may become activated due to local stimulation and additional interaction between receptor/ligand. Step 3: CD11/CD18 mediates firm adhesion. With activated neutrophils and monocytes, a functional upregulation is observed for CD11a, while for CD11b and CD11c both a quantitative and functional upregulation occurs, the upregulation for CD11b being more rapid and important than for CD11c. It is important to note that the functional change of CD11b upon leukocyte activation and/or adhesion can occur despite no measurable increase in CD11b surface expression. The functional change is conformational involving receptor phosphorylation and resulting in increase binding affinity for certain ligands (such as fibrinogen and Factor X) [157]. On the other hand, a quantitative increase in CD11b expression on leukocytes does not imply increased adhesion, unless it is accom- panied by functional change in the receptor [158]. 5.2. Leukocyte activation and biomaterials Contact with cardiovascular devices in vivo activates both neutrophils and monocytes. Indicators of leuko- cyte activation such as l-selectin shedding and CD11b upregulation on leukocytes have been widely observed following angioplasty [159–162], hemodialysis [163–165] and cardiopulmonary bypass [20,22,166–169] (for an extensive review of studies on the expression of leukocyte adhesion molecules with in vitro cardiopul- monary bypass, see Asimakopoulos and Taylor [170]). Degranulation with the release of elastase and lactofer- rin [171–175] and the presence of cytokines [169,176,177], such as IL-1 and TNFa, have been associated with extracorporeal circulation and further demonstrate leukocyte activation. Activation of the respiratory burst is also a common trait with hemodia- lysis [164,178]. Material-induced leukocyte activation also results in increased adhesion. As the biomaterial is larger than a micro-organism and cannot then be engulfed by leukocytes, adherent neutrophils and monocytes under- go a frustrated phagocytosis whereby they release their array of potent oxygen metabolites and proteolytic enzymes [144]. Material characteristics and proteins at the interface appear to modulate the level of activation of adherent leukocytes [153,179,180]. In vivo studies have found activated leukocytes adhering to stents [181,182], oxygenators [183] and hemodialysis mem- branes [184,185]. Material-activated leukocytes also adhere to the endothelium, such as at the anastomoses of a vascular graft or in the lung during extracorporeal circulation. Heparinized human whole blood contact with a PVA hydrogel surface in vitro for 1 h lead to a two-fold upregulation of CD11b, typical of the degree of leukocyte activation induced by phorbol esters [186]. In contrast, whole blood contact with PE and SilasticTM surfaces resulted in minimal CD11b upregulation. We have also reported that many clinical materials can activate isolated neutrophils (without platelets) sus- pended in plasma and that fibrinogen adsorption (plasma but not serum pretreatment) enhances activa- tion [153]. Longer exposures (2 h) to whole blood with polystyrene beads and polystyrene beads grafted with polyethylene glycol (PS-PEG) lead to expression of monocyte TF [187] as well as CD11b upregulation. Activation was dependent on the surface area to volume ratio, but there was no difference in the extent of activation in comparing PS beads with PS-PEG (or PS- PEG-NH2) beads. On the other hand, monocyte TF PRESS terials 25 (2004) 5681–5703 5693 expression and adherent platelet density were all greater on PS than PS-PEG; there were no differences in leukocyte adhesion densities. The mechanisms of leukocyte adhesion on artificial surfaces are not clear, but it appears to be mediated in lize various anticoagulant proteins and activate IN ioma part by the complement product iC3b [72,144]. This is supported by in vitro work showing that inhibition of complement activation in vitro significantly reduced leukocyte adhesion [153,188–190] while fibrinogen also appears to play an important role in leukocyte adhesion to materials [191,192]. In experiments with isolated leukocytes, however, it appears that monocyte TF expression is only partly dependent on complement inhibition but appears to be also dependent on the presence of platelets [64]. The removal of platelets or the blocking of GPIIb/IIIa by monoclonal antibody (7E3) partially lowered the degree of material-induced TF expression while having no effect on CD11b. The presence of platelets on the surface may also mediate leukocyte adhesion via the interaction between P-selectin and PSGL-1 and/or GPIIb/IIIa and CD11b [193,194]. Conflicting reports also exist on the require- ment for platelets in leukocyte adhesion on artificial surfaces [195,196]. The mechanisms of material-induced leukocyte acti- vation as distinct from adhesion also remain unknown. Whether they are directly activated by contact with a foreign surface, via complement activation, kallikrein or platelet activation has not been fully determined. In vitro and in vivo investigations with protease inhibitors [127,175,197–199], complement inhibitors [21,22,186,200–203] and antiplatelet agents [106] suggest that they all play a role, but no one inhibitor has led to consistent results with a significant reduction of material-induced leukocyte activation. Material-induced leukocyte activation may be mediated by several factors and inhibition of one pathway of activation may not be sufficient to result in a significant impact on leukocyte activation. For example, we have shown that complement inhibition via sCR1 was only partially effective in reducing leukocyte activation [186]. On the other hand, a combination of sCR1 and anti- GPIIb/IIIa (which blocks platelet activation) reduced material-induced leukocyte activation to almost back- ground levels. The complexities of leukocyte activation by the PEG modified materials and the inhibitory effects of sCR1 and pentamidine are discussed elsewhere [204]. In summary, the mechanisms that regulate material- induced leukocyte activation are as yet not well under- stood, precluding a clear scientific basis for strategies for a significant reduction in the inflammatory response induced by cardiovascular devices. 5.3. Leukocytes, platelets and coagulation ARTICLE M.B. Gorbet, M.V. Sefton / B5694 Circulating monocytes and neutrophils normally roll on the endothelium. They will however adhere to damaged or stimulated endothelial cells or adherent platelets and further contribute to localized thrombo- genesis. We have evidence [64,187] that similar phenom- platelets. * Association between platelets and neutrophils or monocytes: their interactions may lead to mutual activation and to a microenvironment protected from inhibitors. For example, following blood contact with cardio- pulmonary bypass circuits or ventricular assist devices, TF expression on monocytes has been observed in vitro [205–207] and in vivo [60,61,206]. It has also been shown that CD11b, upregulated on monocytes by cardiopul- monary bypass, was able to directly activate factor X [208] and platelet–leukocyte aggregates have been observed in several scenarios [106,209–211] as noted above. The potential role of leukocytes in thrombogen- esis is underscored by the number of studies that have tried to minimize thrombus formation by the adminis- tration of drugs specifically targeted at leukocytes. Antibodies to block leukocyte adhesion may prove to be a reasonable therapeutic approach in the prevention of thrombus formation as illustrated in in vivo baboon models [212,213]. Overall, however, there is relatively little known on the potential contribution of expression of leukocyte procoagulant activities to thrombogenesis and thrombotic complications associated with the use of biomaterials and cardiovascular devices. 6. Other important factors 6.1. Flow Fluid dynamics affects the growth of thrombi and the deposition of fibrin. The composition difference between arterial and venous thrombi is one old example of this, although the underlying mechanisms are still not well understood. Thorough reviews are available [5,65, 214–216]. Flow determines the rates of transport of cells and proteins to the surface; it can also change the ena apply to biomaterials. The different procoagulant activities of leukocytes may be classified as: * Membrane-associated procoagulant activity: via TF expression on the cell membrane (TF-dependent coagulation pathway) or via TF-independent me- chanisms through factor X binding to CD11b receptors leading to factor Xa generation or fibrino- gen binding to CD11b; or binding of the prothrom- binase complex on the membrane. * Release of procoagulant mediators: degranulation and oxidative products have the capacity to neutra- PRESS terials 25 (2004) 5681–5703 level of receptor expression on platelets and leukocytes. As platelets are an important part of the thrombus, the effect of shear on platelets has been studied extensively. Higher shear results in higher platelet deposition and lower fibrin deposition, while at lower shear the inverse IN ioma is true [214]. High shear, such as the ones observed at stenotic plaques, is also able to induce platelet aggrega- tion even in the absence of any other exogenous factors [217,218]. Conflicting results have been obtained on the effect of flow on leukocyte adhesion while little is known of its effect on leukocyte activation. High shear has been shown to either reduce, increase or leave unchanged leukocyte adhesion on different substrates [195, 219–225]. These conflicting results may be explained by differences in experimental conditions: the presence of red blood cells [220], platelets [195,223,226] and plasma proteins [195]; the surface studied [223] and the state of leukocyte activation [224]. As for the effects of flow on the coagulation cascade, it has been studied less. Current knowledge is limited to Factor Xa generation initiated by the extrinsic pathway and thrombin generation initiated by the intrinsic pathway (with biomaterials). Factor Xa generation by the complex TF:VIIa increases with shear rate (and shear stress) [227,228]. For thrombin generation by the intrinsic pathway, modeling has identified three types of reactions [65]: at low flow, a significant amount of thrombin is produced after a long lag time (over 10 h); at moderate flow, significant thrombin generation is produced in a short time (within minutes); at high flow, low levels of thrombin are produced within seconds. Turbulent flow can be present at anastomoses, joints, and bifurcations of cardiovascular devices and such turbulence also contributes to the observed thrombosis. It is believed to play a significant role in the failure of mechanical heart valves, for example. Turbulent flow (in distinction to recirculation and stagnation zones, which may or may not have the characteristics of turbulence) results in hemolysis and/or cell activation but the mechanisms leading to thrombus formation are still poorly understood. Platelet deposition remains the focus of most studies [229] but the literature on turbulent flow and thrombus formation is more limited. Nonetheless, much effort is done to design devices so that recirculation or stagnation zones are avoided, since these are known niduses for thrombus growth. While the importance of flow has been recognized, our current understanding of its mechanisms is limited mostly to platelets. Many in vitro and in vivo flow models are available and have been successfully used to assess antithrombotic drugs in whole blood [216]. More fundamental research is required on blood coagulation, leukocytes and flow. Previous research with flow has focused on isolated cells or proteins, which is far from the in vivo situation. The critical role of red blood cells in the in vitro study of mechanisms of leukocyte ARTICLE M.B. Gorbet, M.V. Sefton / B adhesion was recently demonstrated by Melder et al. [230]. In the absence of erythrocytes, blocking l-selectin had no effect on lymphocyte adhesion to activated endothelial cells, which was in contradiction with their in vivo observation. Upon addition of erythrocytes to the lymphocytes, l-selectin was then shown to play a significant role in adhesion especially under high shear rate. The use of more physiological experimental conditions (e.g., presence of red blood cells, plasma proteins) should result in significant advances in our knowledge on the effect of mechanical factors on thrombosis and hemostasis. 6.2. Endotoxin Endotoxins, also called LPS, are the component of the outer membrane of gram-negative bacteria and are released into the circulation upon disruption of the intact bacteria (death, cell lysis) [231]. Endotoxin is commonly found everywhere in our environment and it is the most significant pyrogen in parenteral drugs and medical devices. Endotoxins are also present in the digestive system. Their presence in the blood stream may cause septic reactions with a variety of symptoms such as fever, hypotension, nausea, shivering and shock [232]. High concentrations can lead to serious complications such as disseminated intravascular coagulation (DIC), endotoxin shock and adult respiratory distress syn- drome (ARDS). Endotoxins are known to activate complement, the kinin system, leukocytes, platelets and endothelial cells [231,232] and are the ‘‘enemy’’ of both in vitro and in vivo study of blood–material interac- tions. In vivo, they may lead to the complications mentioned above, while in vitro, the presence of this contaminant may affect the results and compromise the conclusions. FDA regulates the acceptable level of endotoxin contamination with medical devices to be 0.5 endotoxin units/ml [233]. There have been few reports of endotoxin contamination with the use of cardiovascular devices. During cardiopulmonary bypass and extracorporeal membrane oxygenation, the presence of endotoxins has been observed in vivo [234,235]. They appear to originate mostly from the gut [236–238] rather than from the materials and are believed to be a reaction to the surgical procedure. During hemodialysis, endotoxin contamination is also an issue and the dialysate is usually the source [232,239]. While endotoxin contam- ination may be present in vivo in some patients and studies, there has been no investigation showing a significant correlation between the magnitude of en- dotoxin contamination and postoperative complications [235,240]. On the other hand, endotoxin contamination during in vitro work may be much more common, as sterile conditions are not always available and the laboratory- PRESS terials 25 (2004) 5681–5703 5695 working environment contributes to their presence. The most conspicuous source of endotoxin may actually be the water since distillation and deionizing columns do not remove endotoxin. Endotoxin has an effect on platelets only at high concentration (over 1 mg/ml on the results. In the orthopedic area, recent studies [258,259] have focussed on endotoxin contamination of cytes; the ability of CD11b to bind Factor X and fibrinogen; the ability of released inflammatory media- tors to activate platelets and block inhibitors of coagulation; and by promoting the association between leukocytes and platelets. In the last 5 years, many leukocyte investigators have discussed the participation of inflammatory cells in coagulation [249–252]. Throm- bosis is viewed now more as a multicellular event rather than just a platelet event [253]. In certain situations, blocking leukocyte contributions to thrombin genera- tion may appear to be a reasonable means to reduce the occurrence of thrombotic complications. Such non- traditional approaches to thrombosis control with biomaterials may be a useful opportunity for further study. The mechanism of biomaterial-associated thrombosis is not fully clear. The role of Factor XII is uncertain while that of TF has not been directly assessed. Both the mechanisms of leukocyte and platelet activation by IN PRESS iomaterials 25 (2004) 5681–5703 microparticles used to assess the effect of wear debris in vitro and have confirmed the significant effect of environmental endotoxin contamination on cytokines. Many washing procedures are now available to ensure endotoxin removal from materials [247,248] and should ensure that the study of blood–material interactions is not impaired by the presence of endotoxin. 7. Conclusions The complexity of blood–material interactions ex- plains our failure to design a material that is entirely blood-compatible. Our current stage of knowledge is far from providing us with a complete mechanism of material-induced thrombin generation. One issue has been the natural scientific tendency to focus on individual aspects of the whole problem rather than considering the various interactions. For example, the biomaterials community has typically looked at platelet interactions in platelet-rich plasma and so is unable to explore interactions between leukocytes and platelets. Alternatively, an anticoagulant prevents thrombin effects from being considered. Of course, without anticoagulants or with whole blood, the experiments get too complicated or sometimes impossible to perform or analyze. Unfortunately, simplifying the system has not allowed us to make real progress. We have also separated thrombosis from its normal equivalent to 5000EU/ml) [241], while leukocytes have been reported to be activated by endotoxin concentra- tions as low as 0.01 ng/ml (equivalent to 0.05 EU/ml) [242,243]. When studying blood–material interactions, endotoxin may be contained in buffers and/or on materials, and its priming and activating effect on leukocytes may affect the observed results. However, it is important to consider that all the studies performed on the effect of endotoxin on leukocyte activation used purified strains of endotoxin while the endotoxin present in laboratory materials and buffers are of an environ- mental nature. Purified endotoxins are much more potent than environmental endotoxins [244] and even among purified endotoxins, their activity might vary [242,245]. Contrary to a purified strain of endotoxin [239], the presence of relatively high levels of environ- mental endotoxins (100 EU/ml) was shown to have little impact on the leukocyte response to hemodialysis [246] in vivo. But when tested in vitro, environmental contamination of a material may have a dramatic effect ARTICLE M.B. Gorbet, M.V. Sefton / B5696 context in inflammation and wound healing. The molecular links between inflammation and thrombosis are undeniable. Inflammation, as characterized by a leukocyte response to a stimulus, may contribute to thrombin generation by the TF expression on mono- materials remain to be further elucidated. As noted above in the context of coagulation, the timing of the events contributing to thrombin formation is also a complex issue. Both Factor XII activation and platelet activation are able to generate thrombin formation within minutes while thrombin generation via leukocyte TF requires hours since TF has to be synthesized. The contribution of leukocyte proteases will also be affected by time since their effect will be dependent on the presence of inhibitors and other inflammatory mediators that can potentiate their action. As illustrated in Fig. 5, the time course of the underlying steps in biomaterial- associated thrombosis may need more consideration Fig. 5. Time scales of biomaterial-associated thrombogenicity. Each component is associated with a different time scale. Protein adsorption and Factor XII activation (assuming it is relevant) occurs within seconds of blood–material contact, producing low levels of thrombin. Platelet activation occurs within minutes creating the phospholipid surface required for assembly of the platelet bound coagulation enzymes and the production of enough thrombin to cause substantial fibrin formation. Leukocyte activation (CD11b upregulation) also occurs within minutes leading to adhesion while TF expression occurs over hours (as transcription and translation must occur first). Complement activation occurs at all these time scales, but whether it is the prelude to leukocyte activation and/or platelet activation is not clear; we suspect that these three are highly interlinked. The interplay between these components will vary depending on the time scale of the situation: a low time constant situation (e.g., high flow, straight tube) would involve different components than a high time constant (e.g., low flow, stagnation zone) situation. Cell 1988;53:505–18. IN ioma [1] Hanson SR. Device thrombosis and thromboembolism. Cardi- ovasc Pathol 1993;2:157S–65S. [2] Bittl JA. Coronary stent occlusion: thrombus horribilis. J Am Coll Cardiol 1996;28:368–70. [3] Clagett GP, Eberhart RC. Artificial devices in clinical practice. In: Colman RW, Hirsh J, Marder VJ, Salzman EW, editors. Hemostasis and thrombosis: basic principles and clinical practice. Philadelphia: Lippincott; 1994. p. 1486–505. [4] Bick RL. Hemostasis defects with cardiac surgery, general surgery and prosthetic devices. In: Bick RL, editor. Disorders of hemostasis and thrombosis. Chicago: American Society of Clinical Pathologist Press; 1992. p. 195–222. [5] Eberhart RC, Clagett CP. Catheter coatings, blood flow, and biocompatibility. Semin Hematol 1991;28:42–8. [6] Edmunds LHJ. Is prosthetic valve thrombogenicity related to design or material? Tex Heart Inst J 1996;23:24–7. [7] Geiser T, Sturzenegger M, Genewein U, Haeberli A, Beer JH. Mechanisms of cerebrovascular events as assessed by procoagulant activity, cerebral emboli and platelet microparti- cles in patients with prosthetic heart valves. Stroke 1998;29: 1770–7. [8] Mickelson JK, Lakkis NM, Villarreal-Levy G, Hughes BJ, Smith CW. Leukocyte activation with platelet adhesion after coronary angioplasty: a mechanism for recurrent disease? J Am Coll Cardiol 1996;28:345–53. [9] Wendel HP, Ziemer G. Coating techniques to improve the hemocompatibility of artificial devices used for extracorporeal circulation. Eur J Cardiothorac Surg 1999;16:342–50. than has hitherto been the case. Experimental studies already have difficulty making the transition from the first few seconds of protein adsorption to the first few minutes of coagulation and cell adhesion. What happens over hours to days as leukocytes synthesise TF and thrombotic deposits become remodeled is almost beyond current experimental capacity. Whether throm- bosis leads to passivation or embolization or some other long-term consequence is still largely unknown. The solution to thrombotic complications associated with cardiovascular devices may not be to try to create a new material that will elicit the proper blood response: the inert cardiovascular biomaterial may be impossible. Rather, more success may be achieved by preventing the adverse effects of a biomaterial by actively blocking the pathway responsible for the inherent thrombogenicity of the materials. Rather than minimizing non-specific biomaterial-associated activation, active inhibition may be the only recourse. Despite more than 50 years of research on blood–material interactions, many ques- tions remain unanswered. The intent of this review was to summarize what we know and to highlight what we have yet to learn. References ARTICLE M.B. Gorbet, M.V. Sefton / B [10] Defraigne JO, Pincemail J, Larbuisson R, Blaffart F, Limet R. Cytokine release and neutrophil activation are not prevented by heparin-coated circuits and aprotinin administration. Ann Thorac Surg 2000;69:1084–91. [11] Moen O, Hogasen K, Fosse E, et al. Attenuation of changes in leukocyte surface markers and complement activation with heparin-coated cardiopulmonary bypass. Ann Thorac Surg 1997;63:105–11. [12] Baufreton C, Jansen PG, Le BP, et al. Heparin coating with aprotinin reduces blood activation during coronary artery operations. Ann Thorac Surg 1997;63:50–6. [13] Muehrcke DD, McCarthy PM, Kottke-Marchant K, et al. Biocompatibility of heparin-coated extracorporeal bypass cir- cuits: a randomized, masked clinical trial. J Thorac Cardiovasc Surg 1996;112:472–83. [14] Gorman RC, Ziats N, Rao AK, et al. Surface-bound heparin fails to reduce thrombin formation during clinical cardiopul- monary bypass. J Thorac Cardiovasc Surg 1996;111:1–11. [15] Redmond JM, Gillinov AM, Stuart RS, et al. Heparin-coated bypass circuits reduce pulmonary injury. Ann Thorac Surg 1993; 56:474–8. [16] Levy M, Hartman AR. Heparin-coated bypass circuits in cardiopulmonary bypass: improved biocompatibility or not. Int J Cardiol 1996;53:S81–7. [17] Fosse E, Thelin S, Svennevig JL, et al. Duraflo II coating of cardiopulmonary bypass circuits reduces complement activation, but does not affect the release of granulocyte enzymes: a European multicentre study. Eur J Cardiothorac Surg 1997;11: 320–7. [18] Videm V, Mollnes TE, Bergh K. Heparin-coated cardiopulmon- ary bypass equipment. II. Mechanisms for reduced complement activation in vivo. J Thorac Cardiovasc Surg 1999;117: 803–9. [19] Janvier G, Baquey C, Roth C, Benillan N, Belisle S, Hardy J-F. Extracorporeal circulation, hemocompatibility and biomaterials. Ann Thorac Surg 1996;62:1926–34. [20] Videm V, Mollnes TE, Fosse E. Heparin-coated cardiopulmon- ary bypass equipment. I. Biocompatibility markers and devel- opment of complications in high risk population. J Thorac Cardiovasc Surg 1999;117:794–802. [21] Gillinov AM, DeValeria PA, Winkelstein JA, et al. Complement inhibition with soluble complement receptor type 1 in cardio- pulmonary bypass. Ann Thorac Surg 1993;55:619–24. [22] Fitch JC, Rollins S, Matis L, et al. Pharmacology and biological efficacy of a recombinant, humanized, single-chain antibody C5 complement inhibitor in patients undergoing coronary artery bypass graft surgery with cardiopulmonary bypass. Circulation 1999;100:2499–506. [23] Murkin JM. Cardiopulmonary bypass and the inflammatory response; a role for serine protease inhibitor. J Cardiothorac Vasc Anesth 1997;11:19–23. [24] Sundaram S, Gikakis N, Hack CE, et al. Nafamostat mesilate, a broad spectrum protease inhibitor, modulates platelet, neutro- phil and contact activation in simulated extracorporeal circula- tion. Thromb Haemost 1996;75:76–82. [25] Rao AK, Sun L, Gorman JH, Edmunds Jr LH. GPIIb/IIIa receptor antagonist Tirofiban inhibits thrombin generation during cardiopulmonary bypass in baboons. Thromb Haemost 1999;82:140–4. [26] Vertress RA, Tao W, Kramer GC, et al. Tumor necrosis factor monoclonal antibody prevents alterations in leukocyte popula- tions during cardiopulmonary bypass. ASAIO J 1994;40: M554–9. [27] Schmaier AH. Contact activation: a revision. Thromb Haemost 1997;77:101–7. [28] Furie B, Furie BC. The molecular basis of blood coagulation. PRESS terials 25 (2004) 5681–5703 5697 [29] Blajchman MA, .Ozge-Anwar AH. The role of the complement system in hemostasis. Prog Hematol 1986;XIV:149–82. [30] Jetsy J, Nemerson Y. The pathways of blood coagulation. In: Beutler E, Lichtman MA, Coller BS, Kipps TJ, editors. Williams hematology. 5th ed. New York: McGraw-Hill; 1995. p. 1227–38. IN ioma [31] Rapaport SI, Rao VM. Initiation and regulation of the tissue factor-dependent blood coagulation. Arterioscler Thromb 1992; 12:1111–21. [32] Morrissey JH, Macik BG, Neuenschwander PF, Comp PC. Quantitation of activated factor VII levels in plasma using a tissue factor mutant selectively deficient in promoting factor VII activation. Blood 1993;81:734–44. [33] Osterud B, Rapaport SI. Activation of Factor IX by the reaction product of tissue factor and Factor VIIa; additional pathway for initiating blood coagulation. Proc Natl Acad Sci USA 1977;74: 5260–4. [34] Lawson JH, Kalafatis M, Stram S, Mann KG. A model for the tissue factor pathway to thrombin. An empirical study. J Biol Chem 1994;269:23357–66. [35] Rand MD, Lock JB, van’t Veer C, Gaffney DP, Mann KG. Blood clotting in minimally altered whole blood. Blood 1996;88: 3432–45. [36] Bauer KA, Rosenberg RD. Control of coagulation factors. In: Beutler E, Lichtman MA, Coller BS, Kipps TJ, editors. Williams hematology. 5th ed. New York: McGraw-Hill; 1995. p. 1239–50. [37] Hirsh J. Oral anticoagulant drugs. N Engl J Med 1991;324: 1865–75. [38] Novotny WF, Girard TJ, Miletich JP, Broze Jr GB. Platelets secrete a coagulation inhibitor functionally and antigenetically similar to the lipoprotein associated coagulation inhibitor. Blood 1988;72:2020–5. [39] McGee MP, Foster S, Wang X. Simultaneous expression of tissue factor and tissue factor pathway inhibitor by human monocytes. A potential mechanism for localized control blood coagulation. J Exp Med 1994;179:1847–54. [40] Cooke ED, Bowcock SA, Lloyd MJ, Pilcher MJ. Intravenous lignocaine in prevention of deep venous thrombosis after elective surgery. The Lancet 1977;ii:797–9. [41] Higuchi DA, Wun TC, Likert KM, Broze GJ. The effect of leukocyte elastase on tissue factor pathway inhibitor. Blood 1992;79:1712–9. [42] Esmon CT. Regulatory mechanisms in hemostasis: natural anticoagulants. In: Colman RWet al,, editor. Hemostasis and thrombosis. Philadelphia: Lippincott; 1997. p. 1597–9. [43] Horbett TA. Principles underlying the role of adsorbed plasma proteins in blood interactions with foreign materials. Cardiovasc Pathol 1993;2:137S–48S. [44] Ziats N, Pankowsky DA, Tierney BP, Ratnoff OD, Anderson JM. Adsorption of Hageman Factor (Factor XII) and other human plasma proteins to biomedical polymers. J Lab Clin Med 1990;116:687–96. [45] Mulzer SR, Brash JL. Identification of plasma proteins adsorbed to hemodialyzers during clinical use. J Biomed Mater Res 1989; 23:1483–504. [46] Cornelius RM, Brash JL. Identification of proteins adsorbed to hemodialyser membranes from heparinized plasma. J Biomater Sci Polym Ed 1993;4:291–304. [47] Matata BM, Courtney JM, Sundaram S, et al. Determination of contact phase activation by the measurement of the activity of supernatant and membrane surface-adsorbed factor XII: its relevance as a useful parameter for the in vitro assessment of haemodialysis membranes. J Biomed Mater Res 1996;31:63–70. [48] Elam J-H, Nygren H. Adsorption of coagulation proteins from whole blood on to polymer materials: relation to platelet activation. Biomaterials 1992;13:3–8. ARTICLE M.B. Gorbet, M.V. Sefton / B5698 [49] Van der Kamp KWHJ, van Oweren W. Factor XII fragment and kallikrein generation in plasma during incubation with bioma- terials. J Biomed Mater Res 1994;28:349–52. [50] Van der Kamp KWHJ, Kauch KD, Feijen J, Horbett TA. Contact activation during incubation of five different polyur- ethanes or glass in plasma. J Biomed Mater Res 1995;29:1303–6. [51] Hong J, Nilsson Ekdahl K, Reynolds H, Larsson R, Nilsson B. A new in vitro model to study interaction between whole blood and biomaterials. Studies of platelet and coagulation activation and the effect of aspirin. Biomaterials 1999;20:603–11. [52] Blezer R, Willems GM, Cahalan PT, Lindhout T. Initiation and propagation of blood coagulation at artificial surfaces studied in a capillary flow reactor. Thromb Haemost 1998;79:296–301. [53] Irvine L, Sundaram S, Courtney JM, Taggart DP, Wheatley DJ, Lowe GDO. Monitoring of Factor XII activity and granulocyte elastase release during cardiopulmonary bypass. ASAIO Trans 1991;XXXVII:569–71. [54] Boisclair MD, Lane DA, Philippou H. Mechanisms of thrombin generation during surgery and cardiopulmonary bypass. Blood 1993;82:3350–7. [55] Boisclair MD, Philippou H, Lane DA. Thrombogenic mechan- isms in the human: fresh insights obtained by immunodiagnostic studies of coagulation markers. Blood Coagul Fibrinolysis 1993;4:1007–21. [56] Burman JF, Chung HI, Lane DA, et al. Role of Factor XII in thrombin generation and fibrinolysis during cardiopulmonary bypass. Lancet 1994;344:1192–3. [57] Colman RW. Mechanisms of thrombus formation. Cardiovasc Pathol 1993;2:23S–31S. [58] Ratner BD. The blood compatibility catastrophe. J Biomed Mater Res 1993;27:283–7. [59] Edmunds Jr LH. Is prosthetic valve thrombogenicity related to design or material? Tex Heart Inst J 1996;23:24–7. [60] Chung JH, Gikakis N, Rao K, Drake TA, Colman RW, Edmunds LHJ. Pericardial blood activates the extrinsic coagula- tion pathway during clinical cardiopulmonary bypass. Circula- tion 1996;93:2014–8. [61] Wilhelm CR, Ristich J, Kormos RL, Wagner WR. Monocyte tissue factor expression and ongoing complement generation in ventricular assist devices patients. Ann Thorac Surg 1998;65: 1071–6. [62] Ernofsson M, Thelin S, Siegbahn A. Monocyte tissue factor expression, cell activation and thrombin formation during cardiopulmonary bypass: a clinical study. J Thorac Cardiovasc Surg 1997;113:576–84. [63] Gorbet MB, Sefton MV. Leukocyte activation and procoagulant activities, in preparation. [64] Gorbet MB, Sefton MV. Material-induced tissue factor expres- sion but not CD11b upregulation depends on the presence of platelets. J Biomed Mater Res 2003;67A:792–800. [65] Basmadjian D, Sefton MV, Baldwin SA. Coagulation on biomaterials in flowing blood: some theoretical considerations. Biomaterials 1997;18:1511–22. [66] Hirsh J, Weitz JI. Antithrombin therapy. In: Califf RM, editor. Acute myocardial infarction and other acute ischemic syn- dromes. St Louis: Mosby; 1996. [67] Hirsh J, Raschke R, Warkentin TE, Dalen JE, Deykin D, Poller L. Heparin: mechanism of action, pharmacokinetics, dosing considerations, monitoring, efficacy and safety. Chest 1995;108: 258S–75S. [68] Rubens FD, Weitz JI, Brash JL, Kinlough-Rathbone RL. The effect of antithrombin III-independent thrombin inhibitors and heparin on fibrin accretion onto fibrin-coated polyethylene. Thromb Haemost 1993;69:130–4. [69] Gustafsson D, Elg M. The pharmacodynamics and pharmaco- kinetcis of the oral direct thrombin inhibitor ximelagatran and PRESS terials 25 (2004) 5681–5703 its active metabolite melagatran: a mini-review. Thromb Res 2003;109:S9–15. [70] Edens RE, Lindhart RJ, Weiler JM. Heparin is not just an anticoagulant anymore: six and one-half decades of studies on the ability of heparin to regulate complement activity. Comple- ment Profiles 1993;1:96–120. IN ioma [71] Misoph M, Schwender S, Babin-Ebell J. Response of the cellular immune system to cardiopulmonary bypass is independent of the applied pump type and of the use of heparin-coated surfaces. Thorac Cardiovasc Surg 1998;46:222–7. [72] Remes R, Williams DF. Immune response in biocompatibility. Biomaterials 1992;13:731–43. [73] Hakim RM. Complement activation by biomaterials. Cardio- vasc Pathol 1993;2:187S–97S. [74] Kazatchkine MD, Carreno M-P. Activation of the complement system at the interface between blood and artificial surfaces. Biomaterials 1988;9:30–5. [75] Johnson RJ. Complement activation during extracorporeal therapy: biochemistry, cell biology and clinical relevance. Nephrol Dial Transplant 1994;9:36–45. [76] Chenoweth DE. Complement activation during hemodialysis: clinical observations, proposed mechanisms, and theoretical implications. Artif Organs 1984;8:281–90. [77] Agostini A, Gardinali M. Complement activation during hemodialysis. J Biomater Appl 1989;4:102–22. [78] Videm V, Fosse E, Mollnes TE, Garred P, Svennevig JL. Time for new concepts about measurement of complement activation by cardiopulmonary bypass? Ann Thorac Surg 1992;54:725–31. [79] Kottke-Marchant K, Anderson JM, Miller KM, Marchant RE, Lazarus H. Vascular graft-associated complement activation and leukocyte adhesion in an artificial circulation. J Biomed Mater Res 1987;21:379–97. [80] Shepard AD, Gelfand JA, Callow AD, O’Donnell Jr TF. Complement activation by synthetic vascular grafts. J Vasc Surg 1984;1:829–38. [81] Johnson RJ. Immunology and the complement system. In: Ratner BD, Hoffman AS, Schoen FJ, Lemons JE, editors. Biomaterials science. San Diego: Academic Press; 1996. p. 173–88. [82] Craddock PR, Fehr J, Dalmasso AP, Brighan KL, Jacob HS. Hemodialysis leukopenia. Pulmonary vascular leukostasis re- sulting from complement activation by dialyzer cellophane membranes. J Clin Invest 1977;59:879–88. [83] Tulunay M, Demiralp S, Tatsan S, et al. Complement (C3, C4) and C-reactive protein responses to cardiopulmonary bypass and protamine administration. Anaesth Intens Care 1993;21:50–5. [84] Lhotta K, Wurzner R, Kronenberg F, Opperman M, Konig P. Rapid activation of the complement system by cuprophane depends on complement component C4. Kidney Int 1998;53: 1044–51. [85] Thylen P, Fernvik E, Lundhal J, Hed J, Jacobson SH. Modulation of CD11b/CD18 on monocytes and neutrophils following hemodialysis membrane interaction in vitro. Int J Artif Organs 1996;19:156–63. [86] Nilsson UR, Larm O, Nilsson B, Storm KE, Elwing H, Nilsson Ekdahl K. Modification of the complement binding properties of polystyrene; effects of end-point heparin attachment. Scand J Immunol 1993;37:349–54. [87] Sefton MV, Gemmell CH, Gorbet MB. What really is blood compatibility. J Biomater Sci Polym Ed 2000;11:1165–82. [88] Gemmell CH. Platelet adhesion onto artificial surfaces: inhibi- tion by benzamidine, pentamidine, and pyridoxal-5-phosphate as demonstrated by flow cytometric quantification of platelet adhesion to microspheres. J Lab Clin Med 1998;131:84–92. [89] Gemmell CH. Activation of platelets by in vitro whole blood contact with materials: increases in microparticle, procoagulant ARTICLE M.B. Gorbet, M.V. Sefton / B activity, and soluble P-selectin blood levels. J Biomater Sci Polym Ed 2001;12(8):933–43. [90] Gemmell CH, Yeo EL, Sefton MV. Flow cytometric analysis of material-induced platelet activation in a canine model: elevated microparticle levels and reduced platelet life span. J Biomed Mater Res 1997;37:176–81. [91] Blockmans D, Deckmyn H, Vermylen J. Platelet activation. Blood Rev 1995;9:143–56. [92] Marcus AJ. Platelet activation. In: Fuster V, Ross R, Topol EJ, editors. Atherosclerosis and coronary artery disease. Philadel- phia: Lippincott-Raven Publishers; 1996. p. 607–37. [93] Rinder HM, Tracey JL, Rinder CS, Leitenberg D, Smith BR. Neutrophil but not monocyte activation inhibits P-selectin- mediated platelet adhesion. Thromb Haemost 1994;72: 750–6. [94] Rinder HM, Bonan JL, Rinder CS, Ault KA, Smith BR. Dynamics of leukocyte-platelet adhesion in whole blood. Blood 1991;78:1730–7. [95] Sims PJ, Wiedmer T. Induction of cellular procoagulant activity by the membrane attack complex of complement. Semin Cell Biol 1995;6:275–82. [96] Jy W, Mao W-W, Horstman LL, Tao J, Ahn YS. Platelet microparticles bind, activate and aggregate neutrophils in vitro. Blood Cells Mol Dis 1995;21:217–31. [97] Siljander P, Carpen O, Lassila R. Platelet derived microparticles associate with fibrin during thrombosis. Blood 1996;87:4651–63. [98] Holme PA, Solum NO, Brosstad F, Pedersen T, Kveine M. Microvesicles bind soluble fibrinogen, adhere to immobilized fibrinogen and coaggregates with platelets. Thromb Haemost 1998;79:389–94. [99] Tans G, Rosing J, Christella M, et al. Comparison of anticoagulant and procoagulant activities of stimulated platelets and platelet-derived microparticles. Blood 1991;77:2641–8. [100] Nieuwland R, Bercmans RJ, Rotteveel-Eijkman RC, et al. Cell- derived microparticles generated in patients during cardiopul- monary bypass are highly procoagulant. Circulation 1997;96: 3534–41. [101] Gemmell CH. Assessment of material-induced procoagulant activity by a modified Russell viper venom coagulation time test. J Biomed Mater Res 1998;42:611–6. [102] Ware JA, Coller BS. Platelet morphology, biochemistry and function. In: Beutler E, Lichtman MA, Coller BS, Kipps TJ, editors. Williams hematology. New York: McGraw-Hill; 1995. p. 1161–225. [103] Calvete JJ. Clues for understanding the structure and function of a prototypic human integrin: the platelet glycoprotein IIb/IIIa complex. Thromb Haemost 1994;72:1–15. [104] van der Wieken LR. Stents and IIb/IIIa receptor blockers combined: usefulness in various types of coronary artery disease. Semin Interv Cardiol 1999;4:77–83. [105] Ellis SG, Bates ER, Schaible T, Weisman HF, Pitt B, Topol EJ. Prospects for the use of antagonists to the platelet glycoprotein IIb/IIIa receptor to prevent postangioplasty restenosis and thrombosis. J Am Coll Cardiol 1991;17:89B–95B. [106] Mickelson JK, Ali MN, Kleiman NS, et al. Chimeric 7E3 Fab (Reopro) decreases detectable CD11b on neutrophils from patients undergoing coronory angioplasty. J Am Coll Cardiol 1999;33:97–106. [107] Coller BS, Anderson KM, Weisman HF. The anti-GPIIb/IIIa agents: fundamental and clinical aspects. Haemostasis 1996;26: 285–93. [108] The EPIC Investigators. Use of a monoclonal antibody directed against the platelet glycoprotein IIb/IIIa receptor in high risk coronary angioplasty. N Engl J Med 1994;330:956–61. [109] Topol EJ, Lincoff AM, Kereiakes DJ, et al. Multi-year follow-up of Abciximab therapy in three randomized, placebo controlled PRESS terials 25 (2004) 5681–5703 5699 trials of percutaneous coronary revascularization. Am J Med 2002;113:1–6. [110] Spinler SA, Hilleman DE, Cheng JWM, et al. New recommen- dations from the 1999 American College of CArdiology/ American Heart Association acute myocardial infarction guide- lines. Ann Pharmacotherapy 2002;35:589–617. IN ioma [111] Cote AV, Berger PB, Holmes Jr DR, Scott CG, Bell MR. Hemorrhagic complications after percutaneous coronary inter- ventions with adjunctive Abciximab. Mayo Clin Proc 2002;76: 890–6. [112] Makoni SN. Acute profound thrombocytopenia following angioplasty: the dilemma in the management and a review of the literature. Heart 2002;86:e18–9. [113] Grunkemeier JM, Tsai WB, Horbett TA. Hemocompatibility of treated polystyrene substrates: contact activation, platelet adhesion, and procoagulant activity of adherent platelets. J Biomed Mater Res 1998;41:657–70. [114] Godo MN, Sefton MV. Characterization of transient platelet contacts on a polyvinyl alcohol hydrogel by video microscopy. Biomaterials 1999;20:1117–26. [115] Sheppard JI, McClung WG, Feuerstein IA. Adherent platelet morphology on adsorbed fibrinogen: effects of protein incuba- tion time and albumin addition. J Biomed Mater Res 1994;28:1175–86. [116] Chinn JA, Ratner BD, Horbett TA. Adsorption of baboon fibrinogen and the adhesion of platelets to a thin film polymer deposited by radio-frequency glow discharge of allylamine. Biomaterials 1992;13:322–32. [117] Hanson SR, Harker LA. Blood coagulation and blood–materials interactions. In: Ratner BD, Hoffman AS, Schoen FJ, Lemons JE, editors. Biomaterials science. San Diego: Academic Press; 1996. p. 193–9. [118] Tsai WB, Grunkemeier JM, Horbett TA. Human plasma fibrinogen adsorption and platelet adhesion to polystyrene. J Biomed Mater Res 1999;44:130–9. [119] Bailly AL, Laurent A, Lu H, et al. Fibrinogen binding and platelet retention: relationship with the thrombogenicity of catheters. J Biomed Mater Res 1996;30:101–8. [120] Gemmell CH, Ramirez SM, Yeo EL, Sefton MV. Platelet activation in whole blood by artificial surfaces: identification of platelet-derived microparticles and activated platelet binding to leukocytes as material-induced activation events. J Lab Clin Med 1995;125:276–87. [121] Cholakis CH, Zingg W, Sefton MV. Effect of heparin-PVA hydrogel on platelets in a chronic canine AV shunt. J Biomed Mater Res 1989;23:417–41. [122] Hanson SR, Harker LA, Ratner BD, Hoffman AS. In vivo evaluation of artificial surfaces with a nonhuman primate model of arterial thrombosis. J Lab Clin Med 1980;95: 289–304. [123] Ip WF, Sefton MV. Platelet consumption by NHLBI reference materials and silastic. J Biomed Mater Res 1991;25:1321–4. [124] Hanson SR, Harker LA. Effects of platelet-modifying drugs on arterial thromboembolism in baboons. Aspirin potentiates the antithrombotic actions of dipyridamole and sulfinpyrazone by mechanisms independent of cyclooxygenase inhibition. J Clin Invest 1999;75:1591–9. [125] Weily HS, Steele PP, Davies H, Pappas G, Genton E. Platelet survival inpatients with substitute heart valves. N Eng J Med 1974;290:534–6. [126] Llanos GR, Sefton MV. Immobilization of poly(ethylene glycol) onto poly(vinyl alcohol) hydrogel: 2. Evaluation of thrombo- genicity. J Biomed Mater Res 1993;27:1383–91. [127] Wachtfogel YT, Bischoff R, Bauer R, et al. Alpha 1-antitrypsin Pittsburgh (Met358-Arg) inhibits the contact pathway of intrinsic coagulation and alters the release of human neutrophil ARTICLE M.B. Gorbet, M.V. Sefton / B5700 elastase during simulated extracorporeal circulation. Thromb Haemost 1994;72:843–7. [128] Simon P, Ang KS, Cam G. Enhanced platelet aggregation and membrane biocompatibility: possible influence on thrombosis and embolism in haemodialysis patients. Nephron 1987;45: 172–3. [129] Hakim RM, Schafer Al. Hemodialysis-associated platelet activation and thrombocytopenia. Am J Med 1985;78:575–80. [130] Peerschke EIB, Ghebrehiwet B. Platelet receptors for the complement component C1q: implication for hemostasis and thrombosis. Immunobiology 1998;199:239–49. [131] Bing DH. Inhibition of guinea pig complement by aromatic amidine and guanidine compounds. J Immunol 1970;105: 1289–91. [132] Andrew JM, Roman Jr DP, Bing DH. Inhibition of four human serine proteases by substituted benzamidines. J Medicinal Chem 1978;21:1202–7. [133] Hauptmann J, Markwardt F. Inhibition of the haemolytic complement activity by derivatives of benzamidine. Biochem Pharmacol 1977;26:325–9. [134] Cox D, Motoyama Y, Seki J, Aoki T, Dohi M, Yoshida K. Pentamidine: a non-peptide GPIIb/IIIa antagonist—in vitro studies on platelets from humans and other species. Thromb Haemost 1992;68:731–6. [135] Cox D, Aoki T, Seki J, Motoyama Y, Yoshida K. Pentamidine is a specific, non-peptide, GPIIb/IIIa antagonist. Thromb Haemost 1996;75:503–9. [136] Allan R, Rodrick M, Knobel HR, Isliker H. Inhibition of the interaction between the complement component C1q and immune complexes. Int Arch Allergy Appl Immunol 1979;58: 140–8. [137] Subbarao K, Kuchubhotla J, Kakkar VV. Pyridoxal 50phosphate—a new physiological inhibitor of blood coagulation and platelet function. Biochem Pharmacol 1979;28:531–4. [138] Takada Y, Arimoto Y, Mineda H, Takada A. Inhibition of the classical and alternative pathways by amino acids and their derivatives. Immunology 1978;34:509–15. [139] Rinder CS, Rinder HM, Smith MJ, et al. Selective blockade of membrane attack complex formation during simulated extra- corporeal circulation inhibits platelet but not leukocyte activa- tion. J Thorac Cardiovasc Surg 1999;118:460–6. [140] Rinder CS, Rinder HM, Smith BR, et al. Blockade of C5a and C5b-9 generation inhibits leukocyte and platelet activation during extracorporeal circulation. J Clin Invest 1995;96: 1564–72. [141] Gorbet MB, Sefton MV. Role of complement and platelets in leukocyte activation induced by polystyrene and PEG-immobi- lized polystyrene beads in whole blood. J Biomed Mater Res, in preparation. [142] Larsson R, Elgue G, Larsson A, Ekdahl KN, Nilsson UR, Nilsson B. Inhibition of complement activation by soluble recombinant CR1 under conditions resembling those in a cardiopulmonary circuit: reduced up-regulation of CD11b and complete abrogation of binding of PMNs to the biomaterial surface. Immunopharmacology 1997;38:119–27. [143] Cassatella MA. The production of cytokines by PMN. Immunol Today 1995;16:21–6. [144] Anderson JM. Mechanisms of inflammation and infection with implanted devices. Cardiovasc Pathol 1993;2:33S–41S. [145] Del Maschio A, Corvazier E, Maillet F, Kazatchkine MD, Maclouf J. Platelet dependent induction and amplification of polymorphonuclear leukocyte lysosomal enzyme release. Br J Haematol 1989;72:329–35. [146] Bazzoni G, Dejana E, Del Maschio A. Platelet-dependent modulation of neutrophil function. Pharmacol Res 1992;26: 269–72. PRESS terials 25 (2004) 5681–5703 [147] Del Maschio A, Dejana E, Bazzoni G. Bidirectional modulation of platelet and polymorphonuclear leukocyte activities. Ann Hematol 1993;67:23–31. [148] Rebuck N, Finn A. Polymorphonuclear granulocyte expression of CD11a/CD18, CD11b/CD18 and l-selectin in normal individuals. FEMS Immunol Med Microbiol 1994;8:189–96. IN ioma [149] Camerer E, Kolsto A-B, Pridz H. Cell biology of tissue factor, the principal initiator of coagulation. Thromb Res 1996;81:1–41. [150] Smith JA. Neutrophils, host defense, and inflammation; a double-edged sword. J Leukocyte Biol 1994;56:686. [151] Kuijipers TW, Hakkert BC, Van Mourik JA, Roos D. Distinct adhesive properties of granulocytes and monocytes to endothe- lial cells under static and stirred conditions. J Immunol 1990; 145:2588–94. [152] van Ginkel CJW, van Haken WG, Oh JIH, Vreeken J. Stimulation of monocyte procoagulant activity by adherence to different surfaces. Br J Haematol 1977;37:35–45. [153] Gorbet MB, Yeo EL, Sefton MV. Flow cytometric study of in vitro neutrophil activation by biomaterials. J Biomed Mater Res 1999;44:289–97. [154] Ginis I, Tauber AI. Activation mechanisms of adherent human neutrophils. Blood 1990;76:1233–9. [155] Jones DA, Smith CW, McIntire LV. Leukocyte adhesion under flow conditions: principles important in tissue engineering. Biomaterials 1996;17:337–47. [156] Butcher EC. Leukocyte-endothelial cell recognition: three or (more) steps to specificity and diversity. Cell 1991;67:1033–6. [157] Arnaout MA. Structure and function of the leukocyte adhesion molecules CD11/CD18. Blood 1990;75:1037–50. [158] Philips MR, Buyon JP, Winchester R, Weissmann G, Abramson SB. Upregulation of the iC3b receptor is neither necessary nor sufficient to promote neutrophil aggregation. J Clin Invest 1988;82:495–501. [159] Takala AJ, Jousela IT, Takkumen OS, et al. Time course of b2 integrin CD11b/CD18 upregulation on neutrophils and mono- cytes after coronary artery bypass grafting. Scand J Thorac Cardiovasc Surg 1996;30:141–8. [160] Serrano CV, Ramires JA, Venturinelli M, et al. Coronary angioplasty results in leukocyte and platelet activation with adhesion molecules expression. J Am Coll Cardiol 1997;29: 1276–83. [161] Inque T, Sakai Y, Morooka S, Hayashi T, Takayanagi K, Takabatake Y. Expression of polymorphonuclear adhesion molecules and its clinical significance in patients treated with percutaneous transluminal coronary angioplasty. J Am Coll Cardiol 1996;28:1127–33. [162] Neumann FJ, Ott I, Gawaz M, Puchner G, Sch .omig A. Neutrophil and platelet activation at balloon-injured coronary artery plaque in patients undergoing angioplasty. J Am Coll Cardiol 1996;27:819–24. [163] Rousseau Y, Carreno M-P, Poignet J-L, Kaztchkine MD, Haeffner-Cavaillon N. Dissociation between complement activa- tion, integrin expression and neutropenia during hemodialysis. Biomaterials 1999;20:1959–67. [164] Cristol JP, Canaud B, Rabesandratana H, Gaillard I, Serre A, Mion C. Enhancement of reactive oxygen species production and cell surface markers expression due to hemodialysis. Nephrol Dial Transplant 1994;9:389–94. [165] Von Appen K, Goolsby C, Mehl P, Goewert R, Ivanovich P. Leukocyte adhesion molecule as biocompatibility markers for hemodialysis membranes. ASAIO J 1994;40:M609–15. [166] Rinder C, Fitch J. Amplification of the inflammatory response: adhesion molecules associated with platelet/white cell responses. J Cardiovasc Pharmacol 1996;27(Suppl. 1):S6–12. [167] El Habbal MH, Carter H Smith L, Elliot MJ, Strobel S. Neutrophil activation in paediatric extracorporeal circuits: effect ARTICLE M.B. Gorbet, M.V. Sefton / B of circulation and temperature variation. Cardiovasc Res 1995;29:102–7. [168] Gillinov AM, Bator JM, Zehr KJ, et al. Neutrophil adhesion molecule expression during cardiopulmonary bypass with bubble and membrane oxygenators. Ann Thorac Surg 1993;56: 847–53. [169] Cameron DE. Initiation of white cell activation during cardiopulmonary bypass: cytokines and receptors. J Cardiovasc Pharmacol 1996;27:S1–5. [170] Asimakopoulos G, Taylor KM. Effects of cardiopulmonary bypass on leukocyte and endothelial adhesion molecules. Ann Thorac Surg 1998;66:2135–44. [171] Horl WH, Riegel W, Steinhauer HB, et al. Granulocyte activation during hemodialysis. Clin Nephrol 1986;26: S30–4. [172] Grooteman MP, van TA, van HA, et al. Hemodialysis-induced degranulation of polymorphonuclear cells: no correlation between membrane markers and degranulation products. Ne- phron 2000;85:267–74. [173] Haag-Weber M, Schollmeyer P, Horl WH. Beta-2-microglobulin and main granulocyte components in hemodialysis patients. Artif Organs 1989;13:92–6. [174] Stegmayr BG, Esbensen K, Gutierrez A, et al. Granulocyte elastase, b thromboglobulin, and C3d during acetate or bicarbonate hemodialysis with hemophan compared to a cellulose acetate membrane. Int J Artif Organs 1992;15:10–8. [175] Wachtfogel YT, Kucich U, Greenplate J, et al. Human neutrophil degranulation during extracorporeal circulation. Blood 1987;69:324–30. [176] Peek GJ, Firmin RK. The inflammatory and coagulative response to prolonged extracorporeal membrane oxygenation. ASAIO J 1999;45:250–63. [177] Frering B, Philip I, Dehoux M, Rolland C, Langlois JM, Desmont JM. Circulating cytokines in patients undergoing normothermic cardiopulmonary bypass. J Thorac Cardiovasc Surg 1994;108:636–41. [178] Himmelfarb J, Lazarus JM, Hakim R. Reactive oxygen species production by monocytes and polymorphonuclear leukocytes during dialysis. Am J Kidney Dis 1991;XVII:271–6. [179] Kaplan SS, Basford RE. Mechanisms of biomaterial induced superoxide release by neutrophils. J Biomed Mater Res 2000;28: 377–86. [180] Kuwahara T, Markert M, Wauters JP. Proteins adsorbed on hemodialysis membranes modulate neutrophil activation. Artif Organs 1989;13:427–31. [181] Serruys PW, Strauss BH, van Beusekom HM, van der Giessen WJ. Stenting of coronary arteries: has a modern Pandora’s box been opened. J Am Coll Cardiol 1991;17:143B–54B. [182] van Beusekom HMM, van der Giessen WJ, Van Suylen RJ, Bos E, Bosman FT, Serruys PW. Histology after stenting of human saphenous vein bypass grafts: observations from surgically excised grafts 3–320 days after stent implantation. J Am Coll Cardiol 1993;21:45–54. [183] Jobes DR. Safety issues in heparin and protamine administration for extracorporeal circulation. J Cardiothorac Vasc Anesth 1998;12:17–20. [184] Grooteman MPC, Bos JC, Van Houte AJ, van Limbeek J, Schoorl M, Nube MJ. Mechanisms of intra-dialyser granulocyte activation: a sequential dialyser elution study. Nephrol Dial Transplant 1997;12:492–9. [185] Castiglione A, Pagliaro P, Romagnoni M, Beccari M, Cavaliere G, Veneroni G. Flow cytometric analysis of leukocytes eluted from haemodialysers. Nephrol Dial Transplant 1991;(Suppl. 2):31–35. [186] Gemmell CH, Black JP, Yeo EL, Sefton MV. Material-induced up-regulation of leukocyte CD11b during whole blood contact: PRESS terials 25 (2004) 5681–5703 5701 material differences and a role for complement. J Biomed Mater Res 1996;32:29–35. [187] Gorbet MB, Sefton MV. Expression of procoagulant activities on Leukocytes following contact with polystyrene and PEG grafted polystyrene beads. J Lab Clin Med 2001;137: 345–55. IN ioma [188] Kao WJ, Sapatnekar S, Hiltner A, Anderson JM. Complement- mediated leukocyte adhesion on poly(etherurethane ureas) under shear stress in vitro. J Biomed Mater Res 1996;32:99–109. [189] McNally AK, Anderson JM. Complement C3 participation in monocyte adhesion to different surfaces. Proc Natl Acad Sci USA 1994;91:10119–23. [190] Cheung AK, Hohnholt M, Gilson J. Adherence of neutrophils to haemodialysis membranes: roles of complement receptors. Kidney Int 1991;40:1123–33. [191] Tang L, Lucas AH, Eaton JW. Inflammatory responses to implanted polymeric biomaterials: role of surface adsorbed immunoglobulin G. J Lab Clin Med 1993;122:292–300. [192] Tang L, Eaton JW. Fibrinogen mediates acute inflammatory responses to biomaterials. J Exp Med 1993;178:2147–56. [193] Eriksson O, Nygren H. Polymorphonuclear leukocytes in coagulating whole blood recognize hydrophilic and hydrophobic titanium surfaces by different adhesion receptors and show different patterns of receptor expression. J Lab Clin Med 2001; 137:296–302. [194] Gemmell CH. Flow cytometric evaluation of material-induced platelet and complement activation. J Biomater Sci Polym Ed 2000;11(11):1197–210. [195] Morley DJ, Feuerstein IA. Adhesion of polymorphonuclear leukocytes to protein-coated and platelet adherent surfaces. Thromb Haemost 1989;62:1023–8. [196] Herzlinger GA, Cumming RD. Role of complement activation in cell adhesion to polymer blood contact surfaces. ASAIO Trans 1980;XXXVI:165–71. [197] Wachtfogel YT, Kettner C, Hack CE, et al. Thrombin and human plasma kallikrein inhibition during simulated extracor- poreal circulation block platelet and neutrophil activation. Thromb Haemost 1998;80:686–91. [198] Wachtfogel YT, Kucich U, Hack CE, et al. Aprotinin inhibits the contact, neutrophil, and platelet activation systems during simulated extracorporeal perfusion. J Thorac Cardiovasc Surg 1993;106:1–9. [199] Himmelfarb J, Holbrook D, McMonagle E. Effects of aprotinin on complement and granulocyte activation during ex vivo hemodialysis. Am J Kidney Dis 1994;24:901–6. [200] Himmelfarb J, McMonagle E, Holbrook D, Toth C. Soluble complement receptor 1 inhibits both complement and granulo- cyte activation during ex vivo hemodialysis. J Lab Clin Med 1995;126:392–400. [201] Gillinov AM, Redmond JM, Winkelstein JA, et al. Complement and neutrophil activation during cardiopulmonary bypass: a study in the complement-deficient dog. Ann Thorac Surg 1994;57:345–52. [202] Finn A, Morgan BP, Rebuck N, et al. Effects of inhibition of complement activation using recombinant soluble receptor 1 on neutrophil CD11b/CD18 and l-selectin expression and release of interleukin-8 and elastase in simulated cardiopulmonary bypass. J Thorac Cardiovasc Surg 1996;111:451–9. [203] Rinder CS, Rinder HM, Johnson K, et al. Role of C3 cleavage in monocyte activation during extracorporeal circulation. Circula- tion 1999;100:553–8. [204] Gorbet MB, Sefton MV. The importance of platelets and complement in material-induced Leukocyte activation in vitro. PhD Thesis, University of Toronton, 2001. [205] Kappelmayer J, Bernabei A, Edmunds Jr LH, Edgington TS, Colman RW. Tissue factor is expressed on monocytes during ARTICLE M.B. Gorbet, M.V. Sefton / B5702 simulated extracorporeal circulation. Circ Res 1993;72: 1075–81. [206] Barstad RM, Hamers MJAG, Moller A-S, Sakariassen KS. Monocyte procoagulant activity induced by adherence to an artificial surface is reduced by end-point immobilized heparin coating of the surface. Thromb Haemost 1998;79:302–5. [207] Stahl RF, Fisher CA, Kucich U, et al. Effects of simulated extracorporeal circulation on human leukocyte elastase release, superoxide generation and procoagulant activity. J Thorac Cardiovasc Surg 1991;101:230–9. [208] Parratt R, Hunt BJ. Direct activation of factor X by monocytes occurs during cardiopulmonary bypass. Br J Haematol 1998;101:40–6. [209] Rinder CS, Bonan JL, Rinder HM, Mathew J, Hines R, Smith BR. Cardiopulmonary bypass induces leukocyte-platelet adhe- sion. Blood 1992;79:1201–5. [210] Gawaz M, Bogner C. Changes in platelet membrane glycopro- teins and platelet leukocyte interactions during hemodialysis. J Clin Invest 1994;72:424–9. [211] May AE, Neumann F-J, Gavaz M, Ott I, Walter H, Sch .omig A. Reduction of monocyte–platelet interaction and monocyte activation in patients receiving antiplatelet therapy after coronary stent implantation. Eur Heart J 1997;18:1913–20. [212] Palabrica T, Lobb R, Furie BC, et al. Leukocyte accumulation promoting fibrin deposition is mediated in vivo by P-selectin on adherent platelets. Nature 1992;359:848–51. [213] Toombs CF, De Graaf GL, Martin JP, Geng JG, Anderson DC, Shebuski RJ. Pretreatment with a blocking antibody to P-sel accelerates pharmacological thrombolysis in a primate model of arterial thrombosis. J Pharmacol Exp Ther 1995;275:941. [214] Turitto VT, Weiss HJ, Baumgartner HR. The effect of shear rate on platelet interaction with subendothelium exposed to citrated human blood. Microvasc Res 1980;19:352–65. [215] Turitto VT, Hall CL. Mechanical factors affecting hemostasis and thrombosis. Thromb Res 1998;92:S25–31. [216] Hanson SR, Sakariassen KS. Blood flow and antithrombotic drug effects. Am Heart J 1998;135:S132–45. [217] O’Brien JR. Shear induced platelet aggregation. Lancet 1990;335:711–3. [218] Oda A, Yokoyama K, Murata M, et al. Protein tyrosine phosphorylation in human platelets during shear-stress induced aggregation (SIPA) is regulated by GPIb/IX as well as GPIIb/ IIIa and requires intact cytoskeleton and endogenous ADP. Thromb Haemost 1995;74:736–42. [219] Kujiper PHM, Gallardo Torres HI, van der Linden JAM, et al. Platelet-dependent primary hemostasis promotes selectin- and integrin-mediated neutrophil adhesion to damaged endothelium under flow conditions. Blood 1996;87:3271–81. [220] Yeo EL, Sheppard J-OI, Feuerstein IA. Role of P-selectin and leukocyte activation in polymorphonuclear cell adhesion to surface adherent activated platelets under physiological shear conditions (an injury vessel wall model). Blood 1994;83: 2498–507. [221] Weber C, Springer TA. Neutrophil accumulation on activated, surface-adherent platelets in flow is mediated by interaction of MAc-1 with fibrinogen bound to aIIbb3 and stimulated by platelet activating factor. J Clin Invest 1997;100: 2085–93. [222] Kujiper PHM, Gallardo Torres HI, Lammers J-WJ, Sixma JJ, Koenderman L, Zwaginga JJ. Platelet and fibrin deposition at the damaged vessel wall: cooperative substrates for neutrophil adhesion under flow conditions. Blood 1997;89:166–75. [223] Bruil A, Sheppard JI, Feijen J, Feuerstein IA. In vitro leukocyte adhesion to modified polyurethane surfaces: III Effects of flow, fluid medium and platelets on PMN adhesion. In: Cooper SL, Bamford CH, Tsuruta T, editors. Polymer biomaterials in PRESS terials 25 (2004) 5681–5703 solution, as interfaces and as solid. Zeist, The Netherlands: VSP; 1995. p. 357–71. [224] Kujiper PHM, Gallardo Torres HI, van der Linden JAM, et al. Neutrophil adhesion to fibrinogen and fibrin under flow conditions is diminished by activation and l-selectin shedding. Blood 1997;89:2131–8. [225] Hernandez MR, Escolar G, Bozzo J, Galan AM, Ordinas A. Inhibition of fibrin deposition on the subendothelium by a [241] Casko G, Suba EA, Elin RJ. Endotoxin-induced platelet activation in human whole blood in vitro. Thromb Haemost ARTICLE IN PRESS M.B. Gorbet, M.V. Sefton / Biomaterials 25 (2004) 5681–5703 5703 CD11b. Studies in a flow system. Haematologica 1997;82: 566–71. [226] Bahra P, Nash GB. Sparsely adherent platelets support capture and immobilization of flowing neutrophils. J Lab Clin Med 1998;132:223–8. [227] Gemmell CH, Nemerson Y, Turitto VT. The effects of shear rate on the enzymatic activity of the tissue factor–factor VIIa complex. Microvasc Res 1990;40:327–40. [228] Nemerson Y, Contino PB. Tissue factor, flow and the initiation of coagulation. In: Seghatchian MJ, Samana MM, Hecker SP, editors. Hypercoagulable states. Boca Raton, FL: CRC Press; 1996. p. 21–8. [229] Bluestein D, Rambod E, Gharib M. Vortex shedding as a mechanism for free emboli formation in mechanical heart valves. J Biomech Eng 2000;122:125–34. [230] Melder RJ, Munn LL, Yamada S, Ohkubo C, Jain RK. Selectin and integrin-mediated T lymphocyte rolling and arrest on TNF- a activated endothelium: augmentation by erythrocytes. Biophys J 1995;69:2131–8. [231] Lynn WA, Golenbock DT. Lipopolysaccharide antagonists. Immunol Today 1992;13:271–6. [232] Lemke HD. Methods for the detection of endotoxins present during extracorporeal circulation. Nephrol Dial Transplant 1994;9:90–5. [233] US Department of Health and Human Services/Public Health Services/Food and Drug Administration. Guideline on valida- tion of the limulus amebocyte lysate test as an end-product endotoxin test for human and animal parental drugs, biological products and medical devices. 1987. p. 1–30. Available at http:// www.fda.gov/cber/gdlns/lal.pdf [234] Hirthler M, Simoni J, Dickson M. Elevated levels of endotoxin, oxygen-derived free radicals and cytokines during extracorporeal membrane oxygenation. J Pediatr Surg 1992;27:1199–202. [235] Nilsson L, Kulander L, Nystr .om S-O, Eriksson O. Endotoxin in cardiopulmonary bypass. J Thorac Cardiovasc Surg 1990; 100:777–80. [236] Butler J, Rocker GM, Westaby S. Inflammatory response to cardiopulmonary bypass. Ann Thorac Surg 1993;55:552–9. [237] Khabar KS, elBarbary MA, Khouqeer F, Devol E, al-Gain S, al- Halees Z. Circulating endotoxin and cytokines after cardiopul- monary bypass: differential correlation with duration of bypass and systemic inflammatory response/multiple organ dysfunction syndrome. Immunol Immunopathol 1997;85:97–103. [238] Rocke DA, Gaffin SL, Wells MT, Koen Y, Brock Utine JG. Endotoxemia associated with cardiopulmonary bypass. J Thorac Cardiovasc Surg 1987;93:832–7. [239] Laude-Sharp M, Caroff M, Simard L, Pusineri C, Kazatchkine MD, Haeffner-Cavaillon N. Induction of IL-1 during hemodia- lysis: transmembrane passage of intact endotoxin (LPS). Kidney Int 1990;38:1089–94. [240] Khazazmi A, Andersen LW, Baek L, Valerius NH, Laub M, Rasmussen JP. Endotoxemia and enhanced generation of oxygen radicals by neutrophils from patients undergoing cardiopulmon- ary bypass. J Thorac Cardiovasc Surg 1989;98:381–5. [242] Weingarten R, Sklar LA, Mathison JC, et al. Interactions of lipopolysaccharide with neutrophils in blood via CD14. J Leukocyte Biol 1993;53:518–24. [243] Wright SD, Ramos RA, Hermanowski-Vosatka A, Rockwell P, Detmers PA. Activation of the adhesive capacity of CR3 on neutrophils by endotoxin: dependence on lipopolysaccharide binding protein and CD14. J Exp Med 1991;173:1281–6. [244] Pearson FC. A comparison of the pyrogenicity of environmental endotoxins and lipopolysaccharides. In: Ten Cate JW, B .uller HR, Sturk A, Levin J, editors. Bacterial endotoxins: structure, biomedical significance, and detection with the Limulus Ame- bocyte Lysate test. New York: Alan Riss, Inc; 1985. p. 251–63. [245] Enders G, Brooks W, Von Jan N, Lehn N, Bayerd .orfer E, Hatz R. Expression of adhesion molecules on human granulocytes after stimulation with Helicobacter pylori membrane proteins: comparison with membrane proteins from other bacteria. Infect Immun 1995;63:2473–7. [246] Tielemans C, Husson C, Schurmans T, et al. Effects of ultrapure and non-sterile dialysate on the inflammatory response during in vitro hemodialysis. Kidney Int 1996;49:236–43. [247] Ragab AA, Van de Motter R, Lavish SA, et al. Measurement and removal of adherent endotoxin from Titanium particles and implant surfaces. J Orthop Res 1999;17:803–9. [248] Hitchins VM, Merritt K. Decontaminating particles exposed to bacterial endotoxin (LPS). J Biomed Mater Res 1999;46:434–7. [249] Gillis S, Furie BC, Furie B. Interactions of neutrophils and coagulation proteins. Semin Hematol 1997;34:336–42. [250] Altieri DC. Inflammatory cell participation in coagulation. Semin Cell Biol 1995;6:269–74. [251] Esmon CT. Introduction: cellular regulation of blood coagula- tion. Semin Cell Biol 1995;6:257–8. [252] May AE, Neumann FJ, Preissner KT. The relevance of blood- cell wall adhesive interactions for vascular thrombotic disease. Thromb Haemost 1999;82:962–70. [253] Marcus AJ. Thrombosis and inflammation as multicellular processes: significance of cell–cell interactions. Semin Hematol 2000;31:261–9. [254] Sengelov H. Complement receptors in neutrophils. Crit Rev Immunol 1995;15:107–31. [255] Pabst MJ. Priming of neutrophils. In: Hellewell PG, Williams TJ, editors. Immunopharmacology of the neutrophil. London: Academic Press; 1994. p. 195–222. [256] Sims PJ. Interaction of platelets with complement system. In: Kunicki TJ, Georges JN, editors. Platelet immunobiology: molecular and clinical aspects. Philadelphia: Lippincott; 1989. p. 354–83. [257] Kirschfink M. Controlling the complement system in inflamma- tion. Immunopharmacology 1997;38:51–62. [258] Hitchins VM, Merritt K. Decontaminating particles exposed to bacterial endotoxin (LPS). J Biomed Mater Res 1999;46:434–7. [259] Ragab AA, Van de Motter R, Lavish SA, Goldberg VM, Ninomiya JT, Carlin CR, Greenfield EM. Measurement and removal of adherent endotoxin from titanium particles and implant surfaces. J Orthop Res 1999;17:803–9. monoclonal antibody to polymorphonuclear leukocyte integrin 1988;59:375–82. Biomaterial-associated thrombosis: roles of coagulation factors, complement, platelets and leukocytes Introduction Coagulation cascade The intrinsic pathway The extrinsic pathway Physiologic inhibitors of coagulation Biomaterials and coagulation pathways Anticoagulants Complement Classical pathway Alternative pathway Regulatory molecules of complement activation Interactions of complement and coagulation cascade Complement activation and biomaterials Platelets Platelet biology Platelets and biomaterials Leukocytes Leukocyte biology Leukocyte activation and biomaterials Leukocytes, platelets and coagulation Other important factors Flow Endotoxin Conclusions References