Various approaches to modify biomaterial surfaces for improving hemocompatibility

https://doi.org/10.1016/j.cis.2004.02.001Get rights and content

Abstract

In this paper, the mechanism of thrombus formation on the surface of polymeric materials and the various approaches of modifying biomaterial surfaces to improve their hemocompatibility are reviewed. Moreover, the blood compatibility of the cellulose membrane grafted with O-butyrylchitosan (OBCS) by using a radiation grafting technique was studied. Surface analysis of grafted cellulose membrane was verified by attenuated total reflection Fourier transform infrared spectroscopy (ATR-FTIR) and electron spectroscopy for chemical analysis (ESCA), which confirmed that OBCS was successfully grafted onto the cellulose membrane surfaces. Blood compatibility of the grafted cellulose membranes was evaluated by platelet rich plasma (PRP) contacting experiments and protein adsorption experiments using blank cellulose membranes as the control. The blood compatibility of OBCS grafted cellulose membranes is better than that of blank cellulose membranes. These results suggest that the photocrosslinkable chitosan developed here has the potential of serving in blood-contacting applications in medical use.

Introduction

In recent years, various biomaterials that are natural or synthetic polymeric materials have been widely used for manufacturing biomedical applications including artificial organs, medical devices and disposable clinical apparatus [1], such as vascular prostheses, blood pumps, artificial kidney, heart valves, pacemaker lead wire insulation, intra-aortic balloon, artificial hearts, dialyzers and plasma separators, which could be used in contact with blood. However, the polymers presently used are conventional materials, such as cellulose, chitosan, poly (tetrafluoroethylene) (PTFE), poly (vinyl chloride) (PVC), segmented polyetherurethane (SPU), polyethylene (PE), silicone rubber (SR), nylon, and polysulfone (PSf). When in direct contact with blood, they are still prone to initiate the formation of clots, as platelets and other components of the blood coagulation system are activated. It is well known that the formation of a thrombus is dependent upon either or both the behaviors of platelets at or near the surface and on the protein-based coagulation cascade, whereas these are harmful for maintaining well-balanced function even life during the treatment of patients.

For example, thrombogenicity of artificial organs, such as artificial heart valves (AHVs), is a serious problem. Many of the deaths in animals given AHVs were caused not by the malfunction of the artificial organ but by blood-clot formation. Due to this problem, patients with mechanical heart valves must undergo lifelong anticoagulation therapy. Even so, the incidence of thromboembolic complications and bleeding complications has been 1.5–3% per year in the USA, and 58% of implanted mechanical heart valves have failed within the past 12 years in China [2]. Improving the hemocompatibility of this kind of device has become a very important task for biomedical material scientists. Similarly, the replacement of conventional hemodialysis treatment for a patient with malfunctioning kidneys by providing him/her with a portable and disposable ‘artificial kidney’ which is connected to implanted artery and venous cannulas on a patient's wrist meets the same problem. During a continuous flow of blood, plasma diffuses through a plasmapheretic membrane, passes through a bed of plasmacompatible sorbent and arrives at an ultrafiltration membrane. Vacuum-operated ultrafiltration removes excess water together with dissolved urea and other small molecules, and sorbent removes medium-sized toxic substances. The hemocompatibility of the sorbent should be improved by chemical modification of the surface [3]. Moreover, cardiovascular pathologies are one of the major causes of death. Such pathologies require in many situations a vascular surgery. Synthetic vascular grafts have been successfully used in the treatment of the pathology of large arteries (internal diameter >6 mm), but the replacement of the smaller sized arteries (internal diameter <6 mm) was often unsuccessful. Synthetic vascular grafts with diameter less than 6 mm are known to be highly thrombogenic and need special treatments to improve their patency after implantation [4].

Thus it can be seen that blood contacting medical devices, which are used not only in living organisms for long periods of time such as the artificial heart and vessels, but also outside of living organisms for short periods of time such as in blood purification devices and blood catheters, must meet the requirement of completely preventing the activation of the coagulation system and clot formation.

In this paper, firstly, we discussed the thrombus formation mechanism on the surface of polymeric materials. As well known, material biocompatibility is generally considered to have close relation with protein adsorption process, because adsorbed proteins may trigger the coagulation sequence. A rapid adsorption of plasma proteins occurs when a foreign material is brought into contact with the blood. The plasma are said to be mainly albumin, γ-globulin, fibrinogen and prothrombin. It has been reported elsewhere that fibrinogen is a major part of the adsorbed protein layer which is partly replaced by kininogen to provide the medium needed for the attachment and, thus, the activation of Factor XII (Hageman Factor) [5]. Factor XII is an intermediate of the intrinsic coagulation pathway and its activation can be initiated by contact with the foreign material or by contact with aggregated platelets. Other pathways involve platelet activation, the complement system and tissue factors. All these aspects of coagulation are inter-related and the blood compatibility assessment of a biomaterial must consider not only the ways in which platelet behavior is affected by contact with the material, but also other blood factors, which might be more subtly influenced by the foreign material [6]. Certainly, the main parameter of blood compatibility commonly investigated is platelet response to the foreign material. Many workers have studied changes in the parameters of platelet activation, e.g. numbers of platelets deposited on a surface either directly or indirectly, clotting test, platelet morphology, assay of platelet release factors such as β-thromboglobulin [7], [8], [9], and test of adenosine nucleotides, serotonin and procoagulatory activators are set free from the now-stimulated platelets. There are suggestions that the release of adenosine nucleotides encourage more platelets to adhere to the surface and the release process of coagulation activators is repeated, which eventually leads to the formation of a platelet aggregate. The process of coagulation starts on the surface where platelets aggregate with the formation of fibrin network. A thrombus is formed from the combination of mutually fused platelets plus the insoluble fibrin and the cells that it has trapped from the blood [5].

There were also some in-depth studies of interfacial charge on the interaction between biomaterials’ surfaces and blood. Hypothesizes concerning the thrombogenicity and the coagulation activation by materials applied in blood contact are briefly summarized: the activation of the blood coagulation by foreign materials is apparently triggered as the level of the kinin system (also designated as the contact system) of the blood plasma. The initial phase of this activation requires negatively charged sites on the material and comprises four proteins: factor XII; factor XI; Kallikrein and high-molecular-weight-kininogen (HMWK) [10]. Numerous negatively charged surfaces and molecules were found to activate the contact system.

Without contact with foreign surfaces, the formation of a platelet thrombus in case of trauma at any point of a blood vessel is apparently induced by the contact of the blood with the subendothelium at the location of the injury. Since platelets are immediately attracted by the injured site over surprisingly far distances from the blood stream, a physical signal has been assumed for the reason [11] acting in combination with the specific interactions between platelet membrane glycoprotein, von Willebrand factor and receptors of the subendothelium. It was found that the intact intima of the vascular system is negatively charged with respect to the adventitia of the vessel but trauma to the vessel is related to the generation of a local positive charge. Therefore positively charged surfaces of materials in contact with blood are suspicious to induce the formation of primary platelet clots. Surfaces with excellent blood compatibility may bear negatively charged groups, e.g. materials derived from proteo-heparansulfate or biomimetic synthetic structures. However, it contradicts current developments that have successfully improved the hemocompatibility using non-ionic or hydrophilic polymers as modifiers. The conclusion drawn from this contradiction is that molecular level information is required with regard to the identification of the charge creating species and their surface concentrations [12].

Much attention has been paid to the thrombus formation mechanism on the surface of polymeric materials for a long time, but there were still many disputes and unknown problems. However, the research tasks of improving hemocompatibility of biomaterials have been carrying out with the development of biomedical requirements. Since the interactions that lead to surface-induced thrombosis occurring at the blood–biomaterial interface become a reason of familiar current complications with grafts therapy, improvement of the blood compatibility of artificial polymer surfaces is, therefore a major issue in biomaterials science. Generally, two main strategies are being followed:

  • 1

    To create surfaces that prevents or suppresses unwanted or uncontrolled reactions of the blood (e.g. activation of the blood coagulation cascade, or activation and aggregation of adherent blood platelets).

  • 2

    To prepare polymers those are inert or passive with respect to blood reactions.

The former is increasingly popular because the interactions between the biological environment and artificial materials are most likely dominated by the materials’ surface properties [13]. It means that the blood compatibility of a material is determined primarily by its surface, rather than by its bulk properties. Further, the blood compatibility can be improved by surface modification, whereas the physical properties of the material remain essentially unchanged.

We would discuss various approaches that were adopted to modify biomaterial surfaces for improving hemocompatibility. In this paper, a categorization of representative methods for polymer surface modification had been suggested as follows:

The methods of chemically immobilizing anticoagulant for enhancing the blood compatibility of polymer are very popular. Various anticoagulants have been devoted to biomedical application. Heparin is an important anticoagulant that is used clinically to minimize thrombous formation on artificial surfaces. It can be directly bound to the surface by insertion of either functional groups or spacer arms, but the activity of heparin was significantly decreased compared with raw heparin [14]. To date, the most successful type of biocompatible surface has been the end-point-attached heparin surface manufactured by Carmeda (Medtronic, Anaheim, CA, USA) [15].

Besides, surface modification with increasing hydrophilicity is believed to be a useful method for improving blood compatibility, and various polymer materials have been modified by water-soluble polymer for biomedical use such as poly(ethylene glycol) (PEG) or poly(ethylene oxide) (PEO) that can prevent plasma protein adsorption, platelet adhesion, and thrombus formation by the steric repulsion mechanism. Steric repulsion by surface-bound water-soluble polymer chains occurs as a result of overlapping polymer layers that could lead to loss in configurational entropy because of volume restriction and/or osmotic repulsion between interdigitated polymer chains. The accepted mechanism for preventing protein adsorption by the grafted PEO chains is that such technique decreased interfacial free energy and the steric repulsion force between PEO chains and the proteins [16]. Similarly, hydrophilic polysulfone membranes, polyvinylpyrrolidone–polysulfone (PVP–PSf), were prepared from PSf membranes covalently conjugated with PVP on the surface. It was found that PVP–PSf membranes gave lower protein adsorption from a plasma solution than PSf membranes. This is also attributed to the hydrophilic surface of the PVP–PSf membranes [17].

Recently, Lin et al. have designed and synthesized novel non-thrombogenic biomaterials by modifying the surface of cellulose with zwitterionic monomer including sulfo ammonium, phosphorous ammonium and carboxylic ammonium. It is considered that the improved antithrombogenicity can be attributed to the zwitterionic structure of the molecules on material surfaces. In aqueous (blood) medium, the molecules with zwitterionic structure not only cannot diffuse into the interior of protein conformation which is mainly maintained by hydrophobic bonds and hydrogen bonds, to affect the synergetic interaction between main chain and side groups, but also can minimize the effect on the exterior ions of conformation, thus favorable to the maintenance of normal conformation of protein and its assembly [18].

Instead of the chemical modification which is generally complicated in the process and often mixed with toxic monomer residues, the physical adsorption method was adopted. For example, the modifier, PEO, was employed and the physical dip-coating method was performed, but the imaginable lack of stability was found. A novel PEO-containing modifier, SPEO-MDI-SPEO (MSPEO), was developed. MSPEO was formed by an amphiphilic coupling-polymer of stearyl poly(ethylene oxide) (SPEO) reacting with 4,4′-diphenylmethane diisocyanate (MDI). When coated onto the outer wall of the intravascular guiding catheters, MSPEO had to be blended with some agents such as Poly(ether urethane) (PEU) that are prone to build a film attaching on the matrix. The results of the blood regular testing were ideal [19]. Amiji and Park also reported that PEO-modified surfaces have been prepared by physical adsorption of PEO homopolymer and block copolymer onto the biomaterial surface for improving blood compatibility [16].

As an anticoagulant, heparin was also employed in the method of physical adsorption. Although lack of stability in immobilization, heparin can maintain its normal conformation and remain highly activity that is useful for improving hemocompatibility.

Surface ozonization is being widely applied in polymer research areas because it has an advantage of uniformly introducing peroxides on the polymer surface even with complicated shape and offers an easy-to-handle, inexpensive method. Moreover, this technique redounds to prepare a satisfactory small-diameter vascular graft (internal diameter <6 mm). When polymer is exposed to ozone gas, peroxides are mainly formed on the carbonyl and carboxyl groups. The generated peroxides are capable of initiating polymerization of vinyl monomers, resulting in polymer grafting onto the ozonated polymeric materials. Lin et al. have reported that silicone film was grafting polymerized with 2-methacryloyloxyethyl phosphorylcholine (MPC) through ozone-induced grafting technology for improving hemocompatibility [20]. Similarly, PSf membranes were treated with ozone to introduce peroxides, and then grafted with either acrylic acid or chitosan, followed by the immobilization of heparin. Blood compatibility was evaluated by Yang and Lin using the activated partial thromboplastin time as well as the adhesion of platelets. The results demonstrated that the blood compatibility of PSf membrane could be improved by conjugating chitosan and heparin [21].

Based on the premise of achieving hemocompatibility through mimicking the chemical constituents of the biologically inert surface of the inactivated platelet membrane, considerable attention has been paid to development of the new technologies that get better hemocompatiblity.

Blood does not coagulate in normal blood vessels. The make-up of the natural blood vessel wall is intriguingly sophisticated and complex. Its structure consists of three layers:

  • 1

    The blood-contacting intima, a monolayer of endothelial cells on a basement membrane.

  • 2

    The media, which are composed of smooth muscle cells embedded in an extracellular matrix of collagen, elastin and mucopolysaccharides.

  • 3

    The adventitia, consisting of fibroblast cells, which are surrounded by an extracellular matrix of mainly collagen.

In fact, this description of the arterial wall also explains why development of functional artificial prostheses for blood vessels meets with so many difficulties [13].

Since the inside surface of normal blood vessels is covered with endothelial cells, endothelial cell-seeding technology is an appreciated method to create a biomimetic micro-environment for the endothelium and for improvement of the functionality of synthetic small-diameter vascular grafts [22], [23]. However, human endothelial cells show little adhesion and no proliferation on the currently available vascular graft materials. Tissue engineering could meet the main requirement for a successful seeding of biomaterial surfaces by a monolayer of endothelial cells to form a scaffold that can facilitate sufficient cell adhesion and promote cell proliferation [24], [25]. It would be interesting to pre-coat these vessels in order to increase their cell adhesive properties. Many results of these researches showed that some polyelectrolyte multilayered films constitute a good interfacial micro-environment at the material surface where those endothelial cells may grow [4].

Moreover, considerable attention has been paid to phospholipids because it is known that they consist of hydrophilic and hydrophobic groups and form the lipid bilayer that are important building units of plasma membranes. If the surface of a polymer possesses a phospholipids-like structure, a significant amount of natural phospholipids in plasma could be adsorbed onto the surface through self-assembly. Phosphorylcholine, which is an electrically neutral and zwitterionic head group that represents the bulk of the phospholipids head groups present on the external surface of blood cells is inert in coagulation assays. There is no doubt that the introduction of the phosphatidylcholine or its analogues into polymer is useful for improving blood compatibility [26]. The blood compatibility that was observed on the surface modified with MPC-polymer derivatives was also confirmed by several research groups [1]. So it could be applied to surface modification of artificial organs and biomedical devices for improving blood and tissue compatibility.

Concepts that the improved blood compatibility can be gotten by grafting the specific proteins, peptides, red blood cells and leucocytes onto material surfaces have been proposed. According to this concept, many research groups have prepared such biomembranes, which have superior thrombo-resistance.

The synthetic methods, such as grafting long alkyl chains or bioactive molecules, often lead to alterations of the original material's physical properties. In contrast, the process of plasma surface modification has been shown to be able to modify the surface properties of a biomaterial without affecting its bulk physical properties. In addition, a lot of chemicals, including those that cannot be polymerized by conventional synthetic methods, can be used to introduce specific functional groups to the substrate. Several studies have indicated that the blood compatibility of small diameter vascular grafts could be improved by plasma deposition processes using various gases [27], [28]. However, due to the complex chemical reactions occurring in the plasma phase and on the substrates of biomaterials, it is difficult to predict the surface chemistry, which is directly related to the biocompatibility. Furthermore, the stability of the plasma-treated surface in air is another concern on the application of plasma treatment [29].

Radiation is widely used to design biochips and in situ photopolymerizable bioadhesives [30]. Radiation is also used in the biomaterials science for surface modification. Biomaterials scientist have taken advantage of the existence of various radiation sources, high-energy electrons, gamma radiation, ultraviolet (UV) and visible light, to manipulate chemical structures of materials in order to improve their biocompatibility. Bengt Rånby has developed ‘surface photografting’, a surface modification processes based on surface grafting reactions initiated by ultraviolet irradiation [31]. These reactions are efficient and fast. They are limited to the surface of the polymers without affecting the bulk properties through giving very thin layers of grafted polymerization of monomers on substrate surface, and bind various materials to surfaces via different surface functional groups.

Because of the complexity and interdependence of the parameters that affect the platelet activation, it is still difficult to evaluate the blood compatibility of a material just from its surface properties. Langmuir–Blodgett (LB) deposition technique is known to be capable of preparing highly ordered monomolecular films with densely packed structure and precisely-controlled thickness. In this technique, the molecules are first spread on water surface and compressed by a movable barrier. This monolayer is then, transferred to a vertical plate at constant surface pressure. Generally, LB film deposition is suitable for amphiphilic molecules and the process is a physical adhesion, contrary to the chemical interaction on the self-assembled monolayers, and thus, the endurance of the film is a concern regarding to the application of LB film. However, the stability of LB films can be improved by cross-linking the molecules after formation. LB technique is used to prepare films of dipalmitoyl phosphatidylcholine (DPPC), dimyrestoyl phosphatidylcholine (DMPC), cholesterol, octadecylamine (ODA), and stearic acid, with thickness of one molecular layer by Lee and Chen who also made comparisons for hemocompatibility [29].

For many biomaterials, however, the surface chemistry is quite complex, and it is difficult to be sure of the chemical composition at the blood-biomaterial interface. The self-assembled monolayer (SAM) deposition has aroused much interest recently. With this technique, the surface properties can be controlled in molecule-level and well defined, highly ordered and orientated surface can be prepared with the goal of developing hemocompatibility of surface.

The interactions between these surfaces and biological molecules have been under investigation. A number of reports have appeared which suggest that cellular adhesion on SAMs can be controlled by the chemistry of the underlying surface. Others have used the control provided by SAMs to direct protein adsorption onto such surfaces. However, the interactions between more complex proteins such as fibronectin and SAMs are still difficult to control, thus resulting in a number of different cellular responses. There was little available imformation regarding the blood compatibility of such surfaces. Moreover, because the films of the SAMs are typically formed by adsorption of alkanethiols onto gold, siliver, or the adsorption of terminally functionalized silanes onto silicon surface, glass, or fused silica, the employment of this technique is limited by the synthesis of terminally functionalized alkanethiols or silanes, as well as the use of restricted substrate such as gold, siliver, silicon surfaces. Recently, formation of SAMs on elastomeric materials such as poly(dimethyl-siloxane) (PDMS) to construct the inner lumen of tubes has been demonstrated by Tegoulia's research group [32].

The technique of surface modifying additives (SMA) was applied by some research groups. It is an example of a low percentage of additives to the bulk polymer, which significantly changes the surfaces properties. Ex-vivo and in vitro tests of cardiopulmonary bypass (CPB) circuits treated with SMA were reported in 1994. SMA technology is based on polysiloxane-containing copolymers that can be blended with base polymer resins before processing. The former co-polymer migrates to the surface during fabrication, creating regions of alternating hydrophilic and hydrophobic domains. This results in a mosaic structure with different charged microdomains on the contact surfaces of CPB. A clinical study of SMA treated CPB resulted in reduced thrombogenicity in SMA treated circuits and inhibited platelet interaction with the biomaterial surface [33]. Defraigne's, Lee's and Van Oeveren's research groups have already reported, respectively, that they adopted the technique of SMA to improve blood compatibility using different modifying additives [34], [35], [36], [37].

Besides the above representative methods, there were still other approaches such as ion beam modification that is a physicochemical surface modification process that results from the impingement of a high-energy ion beam that has been shown to be able to alter and manipulate of the thrombogenicity and control of the cell attachment properties of a material surface without affecting its bulk physical properties, which is adapted to prepare biomaterial surfaces for improve hemocompatibility [38], [39]. In this study, the blood compatibility of the cellulose membranes grafted with O-butyrylchitosan by using a radiation grafting technique was studied.

Many polymer materials have been applied for hemodialysis membranes, such as cellulose, polysulfone, polyacrylonitrile, and nylon [17], [18], [40], [41], [42], [43], [44], [45]. The properties required for a hemodialysis material are excellent ultrafiltration rate, permeability by solutes, mechanical strength, and hemocompatibility.

Cellulose, a polydisperse linear homopolymer composed of regio- and enantioselective β-1,4-glycosidic linked d-glucose unit, is the first abundant natural biopolymer and is readily gotten from a renewable resource, and many of its derives have very broad commercial applications [46], [47]. Although it has good permeability and mechanical strength, its hemocompatibility must be improved further for better hemodialysis [48].

Surface modification is an important technique for improving the blood compatibility of biomedical devices which can not meet the requirement of hemocompatibility when they are used in contact with blood or plasma [49]. Attempts have been made to improve blood compatibility of cellulose by surface modification including immobilization of heparin [50], alkylation with C16-C18 chains to enhance albumin adsorption [51], [52] and graft of PEG [53], [54] and phospholipids polymers [40], [55], [56], [57] onto the membrane surface. In this paper, we chose O-butyrylchitosan (OBCS) to modify cellulose membranes surface due to its special bioactivity.

Chitosan, a (1→4)-linked 2-amino-2-deoxy-β-d-glucan, is prepared by N-deacetylation of chitin, which is the main structural component of crab and shrimp shells. Chitosan has both reactive amino and hydroxyl groups, which can be used to chemically alter its properties under mild reaction conditions. Chitosan exhibits properties that make it desirable candidate for biocompatible and blood-compatible biomaterials [58], [59], [60]. N-acyl chitosans have been reported to be blood compatible materials [61].

By irradiating with ultraviolet light [62], OBCS, a water-soluble derivative of chitosan [63], was covalently immobilized onto the cellulose membrane surfaces using the photosensitive hetero-bifunctional crosslinking reagent, 4-azidobenzoic acid, which was previously bound to OBCS by a reaction between an acid group of the crosslinking reagent and the free amino group of OBCS.

Surface properties of the modified cellulose films were investigated by attenuated total reflection Fourier transform infrared spectroscopy (ATR-FTIR), electron spectroscopy for chemical analysis (ESCA). Moreover, the hemocompatibility of the modified cellulose membranes was evaluated by platelet rich plasma (PRP) contacting experiments and protein adsorption experiments and the results were described.

Section snippets

Materials

Cellulose membranes (Extra cellophane sheets) were purchased from SIGMA-ALDRICH Company, USA. Chitosan powder was obtained from Lianyungang Biologicals Inc., China. Its viscosity average molecular weight was 6.7×105 g/mol, while the degree of deacetylation was 90%.

Platelet rich plasma (PRP) of human blood was supplied by Blood Center of Nanjing Red Cross, China. Bovine fibrinogen (BFG, Sigma, F-8630, USA) was obtained as lyophilized powders.

Other chemicals were reagent grade and used without

Preparation Az-OBCS

OBCS was immobilized covalently on the cellulose membranes surface by using the photosensitive hetero-bifunctional crosslinking reagent, 4-azidobenzoic acid. The reaction scheme for the immobilization is shown in Fig. 1.

Fig. 2a shows the UV spectra of Az-OBCS (water was used as solvent). An absorption at 267 nm, which is assignable to the azidophenyl group, was observed. This result indicates that the azide group had been introduced to the OBCS molecules.

Fig. 2b shows FTIR spectrum of Az-OBCS,

Conclusions

Blood-compatible polymers are important for biomedical applications. It is very necessary to pay attention to the thrombus formation mechanism on the surface of polymeric materials in future for solving unbeknown problems. Since the interactions that lead to surface-induced thrombosis occurs at the blood–biomaterial interface, improvement of the blood compatibility of artificial polymer surfaces is a major issue in biomaterials science. In this study, we have shown that aryl azide with OBCS can

Acknowledgements

The authors wish to thank the financial support provided by the Special Funds for Major State Basic Research Projects of PR China (G1999064705) and the High Technology Research Project of Jiangsu Province (2002HXXSB9B222).

References (67)

  • K. Ishihara et al.

    Biomaterials

    (1999)
  • N. Huang et al.

    Biomaterials

    (2003)
  • V.A. Davankov et al.

    J. Chromatogr. B

    (1997)
  • C. Boura et al.

    Biomaterials

    (2003)
  • G.M. Bernacca et al.

    Biomaterials

    (1998)
  • K. Park et al.

    Biomaterials

    (1990)
  • Y.T. Wachtfogel et al.

    Thromb. Res.

    (1993)
  • C. Werner et al.

    Colloid Surf. A

    (1999)
  • Y.B.J. Aldenhoff et al.

    Biomaterials

    (1997)
  • N. Weber et al.

    Biomaterials

    (2002)
  • M.M. Amiji

    Carbohyd. Polym.

    (1997)
  • A. Higuchi et al.

    Biomaterials

    (2002)
  • J. Zhang et al.

    Colloid Surf. B

    (2003)
  • D.A. Wang et al.

    Biomaterials

    (2001)
  • J.M. Xu et al.

    Colloid Surf. B

    (2003)
  • M.C. Yang et al.

    Polym. Adv. Technol.

    (2003)
  • H. Greisler et al.

    Biomaterials

    (1996)
  • S.L. Goodman et al.

    Biomaterials

    (1996)
  • Y.L. Lee et al.

    Appl. Surf. Sci.

    (2003)
  • R.S. Benson

    Nucl. Instrum. Methods B

    (2002)
  • B. Rånby

    Int. J. Adhes. Adhes.

    (1999)
  • S. Martens et al.

    Cardiovasc. Surg.

    (2003)
  • J.O. Defraigne et al.

    Ann. Thorac. Surg.

    (2000)
  • J.H. Lee et al.

    Biomaterials

    (2000)
  • Y.J. Gu et al.

    Ann. Thorac. Surg.

    (1998)
  • Y. Suzuki

    Nucl. Instrum. Methods B

    (2003)
  • Y. Suzuki et al.

    Nucl. Instrum. Methods B

    (1991)
  • K. Ishihara et al.

    Biomaterials

    (1992)
  • P. Valette et al.

    Biomaterials

    (1999)
  • C.S. Zhao et al.

    Biomaterials

    (2003)
  • N. Moachon et al.

    Biomaterials

    (2002)
  • N.P. Rhodes et al.

    Biomaterials

    (1996)
  • Y.C. Nho et al.

    Radiat. Phys. Chem.

    (2003)
  • Cited by (0)

    View full text