Basic science in vascular surgery Section Editor--Bruce
G e w e r t z , M D , ( C h i c a g o , Illinois)
Interactions at the Blood/Material Interface Howard P. Greisler, MD, Maywood, Illinois
The complex interactions occurring at the blood/material interface of implanted blood-contacting biomaterials largely determine the clinical efficacy of small diameter vascular grafts. Physical characteristics of the material such as chemical composition, electrical charge, surface texture, elastic modulus, and porosity all affect many responses at the interfaces with blood, surrounding tissues, and adjacent artery. Some responses begin instantaneously (deposition of plasma proteins and platelets) while other reactions are somewhat delayed (deposition of leukocytes and monocytes, and migration of endothelial cells and smooth muscle cells). Each of these elements may become activated and release bioactive substances which further affect the behavior of the other entities in this complex and dynamic microenvironment. It is precisely these tissue reactions that regulate thrombogenicity and the development of pseudointimal hyperplasia. (Ann Vasc Surg 1990; 4:98-103) KEY WORDS: Blood/material interface; small diameter vascular grafts; thrombogenicity; pseudointimal hyperplasia.
American Society for Testing and Materials (ASTM) and the Association for the Advancement of Medical Instrumentation (AAMI) have recommended a battery of evaluations of synthetic graft materials. These tests include identification of material (by infrared spectrophotometric or other methods), surface properties (by light microscopy, scanning electron microscopy, surface free energy and critical surface tension), and physical properties such as porosity or permeability and strength including bursting strength, accelerated stress testing, and suture pullout strength. Evaluations of the surface chemistry and texture of vascular grafts involve several techniques. These include infrared spectroscopy using the internal reflection technique, contact angle measurements, scanning electron microscopy, energy dispersive x-ray analysis, surface roughness measurements, surface charge, and determinations of surface potential and Zeta potential. Analysis of protein adsorption involves exposure of the prosthesis either to specific proteins or to
E V A L U A T I O N OF EVENTS OCCURRING AT THE BLOOD/MATERIAL INTERFACE Despite the critical importance of interfacial phenomena in determining the clinical efficacy of an implanted blood-contacting biomaterial, no single test or panel of tests can fully elucidate these complex phenomena. Relevant studies include those designed to characterize the chemical composition, structure and strength of the graft itself, and the texture, composition, and electronegativity of the surface of the graft. Studies of the biologic reactions to the implanted graft include analyses of protein adsorption, and platelet deposition and histologic analyses of tissue reactions at the interface. Vascular graft standards committees of the From the Loyola University Medical Center, Department of Surgery, Maywood, Illinois. Reprint requests: Howard P. Greisler, MD, Loyola University Medical Center, Department of Surgery, 2160 South First Avenue, Maywood, Illinois 60153, 98
VOLUME 4 No 1 - 1990
I N T E R A C T I O N S A T THE BLOOD~MATERIAL I N T E R F A C E
blood with detection of previously radiolabeled proteins. Some of the techniques listed above for surface characterization may also be applied to the grafts following exposure to blood. Platelet deposition is studied by exposure of the graft to blood containing radiolabeled (usually ~l qn-Oxine) platelets in vivo or using in vitro flow loops. Grafts are then removed from flow and counted for gamma irradiation or, while in flow, serially analyzed for adherent radioactivity with a gamma camera. Scanning electron microscopy of grafts once removed from flow can semiquantitatively evaluate the degree of platelet deposition and also show platelet morphology which reflects platelet activation. Platelet activation and degranulation can also be studied by radioimmunoassay for /3-thromboglobulin, platelet factor-4, and other platelet derived substances released following platelet activation. Evaluation of cellular events at the blood/material interface depend predominantly on histologic techniques. In addition, computer assisted planimetry can define percent thrombus free surface area, When Evans blue dye is injected intravenously prior to graft removal, planimetry can define the percent surface coverage by healthy cells (the dye penetrates acellular areas and areas covered by dysfunctional cells). Basic histologic techniques of surface evaluation include light microscopy and SEM. However definitive cell type determination must rest on specific immunocytochemical stains identifying an antigen specific to a particular cell type (eg., Factor VIIIrelated antigen in endothelial cells) or transmission electron microscopic identification of an intracellular organelle specific to a particular cell type (eg., Weibel-Palade bodies in endothelial cells),
HISTOLOGY OF THE INTERFACE Protein adsorption begins almost instantaneously upon establishment of circulation through the graft. Initially this is a function of the concentration of a particular protein in blood and its diffusion coefficient. Thus albumin has the greatest initial interaction with the surface. However because different proteins have different affinities for a particular surface, protein desorption followed by adsorption of a different protein may occur (Fig. I). This dynamic protein/surface interaction is called the Vroman Effect [1,2], and likely affects all other blood/material interactions including those regulating thrombogenicity and vascular healing. Platelet adhesion to the surface begins rapidly and is influenced by the concentration of fibrinogen adsorbed to the surface. Platelet adhesion depends on either receptor-mediated recognition of a binding domain on an adsorbed protein or on a surface induced conformational change or partial denatur-
e t (kd) t
/ / / initial Contact =
soIid surface properties; protein 3-D structure; orientation of protein; hydrophobicity; charge and charge distribution; carbohydrate and sialic acid; concentration; diffusion coefficient;
A d s o r p t i o n after Initial Contact = thermal lability;
denaturability in urea or GdNCI, or as a function of pH; and surface and interfaeial tension as a function of time and temperature,
(neighbors). The ke~, parameters involved in the initial contact phase of adsorption (left ~,ide)and in the time-dependent surface and protein "'denaluralion- processes (right sideJ
Fig. 1. Key parameters involved in initial contact phase of adsorption (left side) and in time-dependent surface and protein "denaturation" processes (right side). (From Andrade ).
ation of a platelet membrane glycoprotein. These events are in turn modulated by surface texture and flow characteristics. Platelet adhesion is often followed by cytoskeletal reorganization and activated platelet degranulation. The bioactive substances within the dense bodies and alpha granules are released which results in a yet greater platelet response. Platelet release products include thromboxane A 2 which results in vasoconstriction, platelet aggregation, and neutrophil adhesion. Platelet factor-4 increases platelet aggregation and inhibits circulating proteases. Platelet derived growth factor (PDGF) is chemotactic and mitogenic for fibroblasts and smooth muscle cells and may be involved in pseudointimal hyperplasia. Coincident with platelet release is the activation of coagulation pathways and thrombin generation which further potentiates platelet aggregation, Biomaterials may activate both classical and alternative complement pathways resulting in C5a generation. The extent of complement activation depends upon the material used; for example, Dacron elicits greater activation than ePTFE . Generated C5a is strongly chemotactic for monocytes which likely play important roles in regulating other tissue reactions. Circulating polymorphonuclear leukocytes (PMNL) are deposited on the blood/material interface in response to chemoattractants including C5a and leukotriene B 4 (LTB4). Products of PMNL including products of oxygen metabolism may retard reendothelialization. Neutrophils also adhere to the endothelial cells growing across the anasto-
I N T E R A C T I O N S A T THE B L O O D ~ M A T E R I A L I N T E R F A C E
mosis. Neutrophils will adhere preferentially to cultured endothelial cells as compared to other cell lines  consistent with observations of neutrophil margination in capillaries. In addition to products of oxygen metabolism, activated neutrophils produce proteases including collagenase and elastase which may inhibit inner capsular tissue incorporation. Circulating monocytes adhere to the surface and infiltrate into the inner capsule, differentiating into activated macrophages. These macrophages are capable of synthesizing many bioactive substances which may play a central role in regulating the healing of vascular grafts. Monocyte recruitment to the vascular graft is highly complex and is stimulated by products of plasma, of other ingrowing cells, and of the graft and the fibrin coagulum as well as the endothelial cells migrating across the anastomosis. PDGF is chemotactic for monocytes . Endothelial cells produce PDGF which is released in increased amounts in vitro when these cells are injured by endotoxin or phorbol esters . Clowes  reported a persistently increased mitotic index in the endothelial population within the pannus ingrowth area suggesting chronic cell injury resulting in continued release of PDGF and persistent monocyte recruitment. Other important monocyte recruitment factors include IL-1 (produced by macrophages among other cells), LTB 4, and platelet factor-4. Plasma-derived monocyte recruitment factors include the complement derived peptide C5a and thrombin. Extracellular matrix products chemotactic for monocytes include fragments of collagen, elastin, and fibronectin. Once present the monocyte may differentiate into an activated macrophage. Our laboratory has demonstrated that the chemical composition of biomateriats may differentially activate the macrophage [8-10] thereby potentially altering the release of macrophage-derived bioactive substances. Humans differ from all other species in their apparent inability to reendothelialize the surface of vascular grafts. Pannus ingrowth results in endothelium residing within 1 to 2 cm of either anastomosis. The other potential sources of endothelium, transinterstitial ingrowth across the prosthetic wall or fallout of circulating cells onto the surface, have not yet been identified in man. Clowes and Reidy have evaluated the process of spontaneous reendothelialization of vascular grafts in the baboon. Using standard (30 micron internodal distance) ePTFE grafts, endothelialization derived from the cut edges and involved both migration and proliferation . However, even at one year, the central portions of 9 cm long grafts were not endothelialized. In contrast, complete reendothelialization occurred when 60 micron internodal distance ePTFE grafts were implanted into the same model and this endothelialization apparently derived from transmural capillary ingrowth [ 11]. These investiga-
ANNALS OF VASCULAR SURGERY
tors found persistent elevations in endothelial cell turnover despite complete reendothelialization suggesting chronic cell injury. Interestingly, smooth muscle proliferation and intimai thickening occurred in those areas with an apparently confluent endothelium suggesting a possible link between active endothelial cell (a source of PDGF) and intimal thickening. Our laboratory has also reported reendothelialization of the surfaces of implanted bioresorbable lactide/glycolide prostheses through transinterstitial capillary ingrowth resulting in islands of ingrowing endothelial cells [12,13].
SIGNALS AT THE INTERFACE initially the characteristics of the prosthesis including surface texture, electronegativity, hydrophobicity, and surface tension all contribute to the Vroman Effect. The long term consequences are likely great when considering that the relative concentrations of different proteins modulate platelet deposition, thrombogenicity, inflammatory processes, monocyte and neutrophil adhesion, and endothelial cell and smooth muscle cell regeneration. Platelet deposition follows and is influenced by protein adsorption. A variety of methods have been used to passivate surfaces. However some degree of platelet deposition still occurs with all of these methods. Surface passivation also raises the theoretical concern that attempts to decrease platelet adhesion may simultaneously increase the risk of microembolization. Platelet deposition is influenced by shear rate and, indirectly, by graft compliance which affects the flow rate of blood during diastole. Flow reversal or flow stagnation in diastole increases the exposure time of circulating elements to the graft surface. Conditions of non-laminar flow may also produce areas of flow separation leading to increased particle residence time. Such characteristics are prominent in the concavities of crimped grafts. Endothelial cells are very much influenced by local hemodynamics. Endothelial cells align themselves in the direction of flow when that flow is laminar. However, turbulent flow patterns alone may result in DNA synthesis without cellular realignment, retraction or loss . Function of endothelial cells is also affected by hemodynamics. Fluctuating shear stresses may increase endothelial cell pinocytosis  which may be relevant to the transendothelial transport of macromolecules in the area of pannus ingrowth and thus may affect pseudointimal hyperplasia. In addition endothelial cell production of prostacyclin is directly related to shear stress . Biomechanical and hemodynamic factors also affect smooth muscle cell activities. Cyclic deformation of smooth muscle cells in vitro results in
I N T E R A C T I O N S A T THE BLOOD~MATERIAL I N I ~ R F A C E
VOLUME 4 No 1 - 1990
METABOUSt~ [ FORt.~1101~
\~Rop~c.E MOL~U~..ES PROUFF.J~N OF
Fig. 2. Sequence of events in monocyte recruitment and activation and roles of activated macrophage in vascular healing.
their increased synthesis of collagen, hyaluronate, and chondroitin-6-sulfate , potentially relevant to the issue of pseudointimal hyperplasia.
C E L L U L A R RESPONSES--MEDIATORS OF HEALING AT THE INTERFACE Once platelets, blood cells, and mesenchymal tissues are deposited at the surface, they may become activated and release bioactive substances which regulate healing. Platelets, macrophages, endothelial cells, and smooth muscle cells are all capable of producing growth factors. Growth factors can be divided into competence factors and progression factors. Competence factors stimulate the entrance into the S phase (DNA synthesis) only in the presence of additional factors found in serum; progression factors stimulate passage into the S phase without other agents being present. When activated platelets release their granule contents, several growth factors including PDGF, epidermal growth factor (EGF), and transforming growth factor/3(TGF-/3) are released. In addition to these growth factors, platelets may mediate vascular healing through release of PF-4 and/3-TG. PDGF is a cationic (pI 9.5 to 10.4) protein strongly mitogenic for fibroblasts and smooth muscle cells. PDGF is stored in the alpha granules of platelets and consists of two polypeptide chains of molecular weights 14,400 and 17,000 which are held together by disulfide bonds. PDGF binds with high affinity to smooth muscle cells and fibroblasts but not to endothelial cells which lack the PDGF receptor. In addition, PDGF is a chemoattractant for smooth muscle cells, neutrophils, and monocytes. Inasmuch as PDGF has a half life in circulation of less than two minutes, it is likely to exert an effect only on cells in the local microenvironment. Once receptor bound, PDGF induces a variety of metabolic events prior to cell division. These events
include activities related to arachidonic acid metabolism, cycloxygenase production, and enhancement of the binding of low density lipoproteins and increasing cholesterol synthesis. PDGF is also a potent vasoconstrictor . Only a brief exposure of cultured cells to PDGF is required for PDGF induction of cell competency to progress through G0/Gj into the S phase of DNA synthesis . The role of platelet derived PDGF in vivo on thromboresistance and on pseudointimal hyperplasia and atherogenesis remains speculative and in vivo studies to date have been difficult to interpret. Platelets also release EGF, another mesenchymal cell mitogen active against some endothelial cell and vascular smooth muscle cell lines. The role of platelet EGF in vivo on vascular healing in unknown. TGF-/3 released from platelets may also affect certain healing characteristics at the blood/material interface. TGF-/3 may stimulate proliferation of fibroblast lines or inhibit proliferative activity in other fibroblast cultures depending upon experimental conditions. Numerous in vivo and in vitro studies have demonstrated that activated macrophages release products affecting endothelial cell, smooth muscle cell, and fibroblast proliferation. Release of these growth factors is apparently dependent upon activation of the macrophage which is frequently related to the phagocytosis of a variety of test materials. We now know that macrophage, demonstrated by Shimokado , may predispose to pseudointimal hyperplasia. Production of IL- 1 [ 2 1 ] may be relevant to thrombogenicity inasmuch as IL-I may enhance endothelial cell procoagulant activities and may be also stimulate fibroblast proliferation which may be an affect mediated by IL = 1 induction of PDGF transcription by the fibroblasts . Basic FGF production by macrophages  may enhance proliferative activity of endothelial cells. In addition, macrophages synthesize and secrete other bioactive substances which are involved in regulating blood/material interactions. Macrophages release products of oxygen metabolism that may result in endothelial cell injury. Macrophages also produce coUagenase, elastase and complement components, all of which may mediate inner capsule cellular injury. Activated macrophages produce bioactive lipids including products of arachidonate metabolism (which may affect vascular tone and platelet adherence) as well as LTB 4 (which may be chemotactic for neutrophils and monocytes). Different biomaterials may differentially attract and activate the macrophage to release bioactive substances. Our laboratory has utilized bioresorbable lactide/glycolide prostheses which are phagocytized by macrophages and result in an enhanced transinterstitial ingrowth of capillaries, endothelial
INTERACTIONS A T THE BLOOD~MATERIAL INTERFACE
O-o Control z~ - - - ~ ~ c r o n
4 0 0 I~'~
/ \ /
~ ........ o
hemostasis and thrombosis and the release of PDGF from cultured endothelial cells. In contrast, Castelott has shown inhibition of smooth muscle cell proliferation in response to a heparin-like substance produced by cultured confluent endothelial cells .
MACROPHAGE CONDITIONED MEDIA L E - I I CELL [ 3 H I THYMIDINE UPTAKE ~
Supported in part by National Institutes of Health Grant No ROI HL41272.
Weeks Fig. 3. Tritiated thymidine incorporation into newly synthesized DNA by previously quiescent murine capillary lung endothelial (LE-II) cells after 16-hour exposure to 1:10 dilution of media conditioned by macrophages cultured in presence of PG910, Dacron, or neither during one week periods after five to 10 weeks in culture. Expressed as percent increase above quiescent levels. (From Greisler ).
cells, and myofibroblasts into the inner capsules [12,13,24-26]. Concurrent in vitro studies evaluating macrophage biomaterial interactions have demonstrated that macrophages cultured with these bioresorbable materials release into their media significantly greater concentrations of substances capable of increasing DNA synthesis in quiescent endothelial cells [8,9] (Fig. 3). Endothelial cells on the blood/material interface derived either from pannus ingrowth, transinterstitial ingrowth, or endothelial cell seeding techniques may play a variety of roles in thrombogenicity, vascular healing, and vascular tone. Endothelial cells regulate transport of nondiffusible plasma proteins and tipoproteins into the subendothelial space. Endothelial cell surface LDL receptors result in LDL uptake and acetylation of LDL, making it available to macrophages and smooth muscle cells. Endothelial cells express both anticoagulant and procoagutant activity. Endothelial cell-derived growth factors are capable of stimulating smooth muscle cell and fibroblast proliferation while endothelial cell derived glycosaminoglycans may inhibit smooth muscle cell proliferation. Endothelial cells may also influence vascular tone by secretion of endothelial cell derived relaxant factor (EDRF) and endothelin (a vasoconstrictor). A significant portion of endothelial cell derived growth factor activity is PDGF . Endothelial cell derived PDGF is released in vitro at a constant rate which can be increased by cell injury . This PDGF release can also be suppressed by administration of acetylated LDL . Recent studies have also demonstrated links between the processes of
REFERENCES 1. ANDRADE JD, HLADY V. Plasma protein adsorption: the big twelve. Ann N Y Acad Sci 1987;516:158-172. 2. VROMAN L. What time does the next protein arrive? Trans Soc Biomater 1986;9:59. 3. SHEPARD AD, GELFAND JA, CALLOW AD, O ' D O N N E L L TF Jr. Complement activation by synthetic vascular prostheses. J Vasc Surg 1984;1:829438. 4. LACKIE JM, DeBONO D. Interactions of neutrophil granulocytes and endothelium in vitro. Microvasc Res 1977; 13:107-112. 5. DEUEL TF, SENIOR RM, HUANG JS, GRIFFIN GL. Chemotaxis of monocytes and neutrophils to platelet-derived growth factor. J Clin Invest 1982;69:1046-1049. 6. FOX PL. DICORLETO PE. Regulation of production of a platelet-derived growth factor-like protein by cultured bovine aortic endothelial cells. J Cell Physiol 1984;121:298308. 7. CLOWES AW, KIRKMAN TR, CLOWES MM. Mechanisms of arterial graft failure. II. Chronic endothelial and smooth muscle cell proliferation in healing polytetrafluoroethylene prostheses. J Vasc Surg 1986;3:877-884. 8. GREISLER HP, DENNIS JW, ENDEAN ED, ELLINGER J, FRIESEL R, BURGESS WH. Macrophage/biomaterial interaction: the stimulation of endothelialization. J Vasc Surg 1989;9:588-593. 9. GREISLER HP. Small diameter vascular prostheses: macrophage-biomaterial interactions with bioresorbable vascular prostheses. Trans ASAIO 1988;XXXIV:1051-1059. 10. GREISLER HP. Pharmacology of the arterial wall. In: WHITE RA (ed). Atherosclerosis: Human pathology and experimental methods and models. Boca Raton; CRC Press, 1989, pp. 111-150. I 1. CLOWES AW, KIRKMAN TR, REIDY MA. Mechanisms of arterial graft healing. Am J Patho! 1986;123:220-230. 12. GREISLER HP, ELLINGER J, SCHWARCZ TH, GOLAN J, RAYMOND RM, KIM DU. Arterial regeneration over polydioxanone prostheses in the rabbit. Arch Surg 1987; 122:715-721. 13. GREISLER HP, DENNIS JW, ENDEAN ED, KIM DU. Derivation of neointima of vascular grafts. Circulation 1988; 78(Suppl /):16-I12. 14. DAVIES PF, REMUZZI A, GORDON EJ, DEWEY CR Jr, GIMBRONE MA Jr. Turbulent fluid shear stress induces vascular endothelial cell turnover in vitro. Proc Nat Acad Sci USA 1986;83:2114-2117. 15. DAVIES PF, DEWEY CR Jr, BUSSOLARI SR, GORDON EJ, GIMBRONE MA Jr. Influence of hemodynamic forces on vascular endothelial function. In vitro studies of shear stress and pinocytosis in bovine aortic endothelial cells. J Clin Invest 1984;73:1121-1129. t6. FRANGOS JA, MCINTIRE LV, ESKIN SG, IVES CL.
VOLUME 4 No 1 - 1990
INTERACTIONS A T THE BLOOD/MA TERIAL INTERFACE
Flow effects on prostacyclin production by cultured human endothelial cells. Science 1985;227:1477-1479. LEUNG DYM, GLAGOV S, MATHEWS MB. Cyclic stretching stimulates synthesis of matrix components by arterial smooth muscle cells in vitro. Science 1976:191: 475~,77. BERK BC, ALEXANDER RW, BROCK TA, GIMBRONE MA Jr, WEBB RC. Vasoconstriction: a new activity for platelet-derived growth factor. Science 1986;232:87-90. HELDIN C-H, RONNSTRAND L. Characterization of the receptor for platetet derived growth factor on human fibroblasts: demonstration of an intimate relationship with a 185,000-dalton substrate for the platelet-derived growth factor-stimulated kinase. J Biol Chem 1983;258:10054-10061. SHIMOKADO K, RAINES EW, MADTES DK, BARRETT TB, BENDITT EP, ROSS R. A significant part of macrophage-derived growth factor consists of at least two forms of PDGF. Cell 1985;43:277-286. SCHMIDT JA, MIZEL SB, COHEN D, GREEN 1. Interleukin 1, a potential regulator of fibroblast proliferation. J lmmunol 1982;128:2177-2182. RAINES EW, DOWER SK, ROSS R. Interteukin-I mitogenic activity for fibroblasts and smooth muscle cells is due to PDGF-AA. Science I989:243:393-396.
23. BAIRD A, MORMEDE P, BOHLEN P. Immunoreactive fibroblast growth factor in cells of peritoneal exudate suggests its identity with macrophage-derived growth factor. Biochem Biophys Res Commun 1985;126:358-364. 24. GREISLER HP+ Arterial regeneration over absorbable prostheses. Arch Surg 1982;117:1425-1431. 25. GREISLER HP, ELLINGER J, SCHWARCZ TH, GOLAN J, RAYMOND RM, KIM DU. Arterial regeneration over polydioxanone prostheses in the rabbit. Arch Surg 1987; 122:715-721. 26. GREISLER HP, ENDEAN ED, KLOSAK J J, ELLINGER J, DENNIS JW, BUTTLE K, KIM DU. Potyglactin 910/ polydioxanone bicomponent totally resorbable vascular prostheses. J Vasc Surg t988;7:697-705. 27. DICORLETO PE, BOWEN-POPE DF. Cultured endothelial cells produce a ptatelet derived growth factorqike protein. Proc Natl Acad Sci USA 1983;80:191%1923. 28. FOX PL, DICORLETO PE. Acetylated tow density lipoprorein suppresses production of platelet-derived growth factor by cultured endothelial cells. J Cell B/o/ t985;101:107a. 29. CASTELLOT JJ Jr, ADDONIZIO ML, ROSENBERG R, KARNOVSKY MJ. Cultured endothelial cells produce a heparin-like inhibitor of smooth muscle cell growth. J Cell Biol 1981 ;90:372-379.