Secondary Logo

Journal Logo

Tissue Engineering

In Vivo Evaluation of Hydroxyapatite Nanocoating on Polyester Artificial Vascular Grafts and Possibility as Soft-Tissue Compatible Material

Kadono, Hiroyuki*†‡; Furuzono, Tsutomu; Masuda, Miwa; Okada, Masahiro; Ueki, Mitsuki; Takamizawa, Keiichi; Tanaka, Ryoichi§; Miyatake, Kunio; Koyama, Yoshihisa*; Takakuda, Kazuo*

Author Information
doi: 10.1097/MAT.0b013e3181c945ae
  • Free


Artificial vascular grafts are one of the most useful prostheses in cardiac and vascular surgery. The clinical outcomes of these grafts give beneficial effects, but there are still many concerns, including tissue irritation and foreign body reaction.1 Grafts made of woven fabric have sometimes been presealed using albumin1 or gelatin to prevent perivascular blood leakage. By such processing, improvement of the blood compatibility of the polyester graft is also expected.2 However, there are concerns about inflammatory response with respect to sealed artificial grafts. The use of crosslinking agents, such as glutaraldehyde or formaldehyde, which are used to crosslink albumin or gelatin, involves the risk of evoking cytotoxic reactions.3 Furthermore, these biological materials increase the possible risk of infection, such as bovine spongiform encephalopathy (BSE).

Recently, we developed a polyester graft coated covalently with sintered hydroxyapatite (HAp) nanocrystals4 instead of animal derivative reagents, such as albumin or gelatin, to increase blood compatibility and biological safety. Generally, HAp has been used as hard tissue-compatible material, because it bonds directly to bone when implanted.5 If bioceramics can be incorporated with such devices as artificial vascular grafts, their clinical usefulness will become higher in accordance with the improved soft-tissue compatibility. Aoki and coworkers6 first reported on the challenging in vivo research of using artificial vascular grafts made of rigid HAp ceramic tubes. However, the brittleness of the materials hinders further investigation. In contrast to such a rigid vascular graft, our HAp nanocrystal coating, achieved by “soft nanoceramic processing,” is a novel HAp coating method, which can coat with the nanocrystals covalently linked on polymer surfaces several tens of nanometers in thickness with only a small change in the elastic modulus of the polyester.4 The material coated by this process showed good cell adhesiveness with several cell lines.4

In this article, HAp nanocrystal-coated polyester grafts were prepared according to our previous literature, and a diameter compliance measurement of the graft was conducted. The grafts were implanted in the bilateral common carotid arteries of mongrel dogs for 12 weeks and histologic observation was performed around the inner lumen of the modified grafts. The effectiveness of the HAp nanocrystal coating in the vascular system is discussed on the basis of the histologic evaluation.

Materials and Methods


The artificial vascular grafts (made of polyester fabrics; internal diameter, 5 mm; wall thickness, 170 μm; length, 20, 30, or 50 mm; water permeability 50–150 ml·min−1·cm−2 at 120 mm Hg and 37°C) were kindly donated by Ube Industries, Ltd., Tokyo, Japan. γ-Methacryloxypropyl triethoxysilane (MPTS) was donated by Shin-Etsu Chemical Industries Co., Tokyo, Japan. Benzyl alcohol (guaranteed reagent; Nacalai Tesque Inc., Kyoto, Japan), methanol (superior quality of reagent; WAKO Pure Chemicals Co. Ltd., Osaka, Japan), and 3% H2O2 (superior quality of reagent; SIGMA, Tokyo, Japan) were used without further purification. Water was purified with a Milli-Q system (Millipore Corp., Bedford, MA). The dispersed HAp nanocrystals with an average diameter of 50 nm and spherical-like shape were prepared with a wet chemical process and used after calcination with an antisintering agent—poly(acrylic acid, calcium salt)—at 800°C for 1 hour, as described in our previous reports.7–9

HAp Nanocrystal Coating on Vascular Graft

The HAp nanocrystal-coating on the polyester vascular graft was conducted according to a previous report.4 First, the polyester graft was carefully immersed in a 0.2 N aqueous NaOH solution at 60°C for 30 min and then rinsed with Milli-Q water to generate hydroxyl groups and carboxyl groups on the surface. Next, graft polymerization of MPTS on the alkali-treated graft was conducted with benzyl alcohol used as the medium and with H2O2 used as the initiator10 under N2 at 80°C for 120 min. After the graft polymerization, the poly(MPTS)-grafted polyester graft was washed with ethanol several times and then dried under reduced pressure for 30 min at 70°C. It was then soaked in the HAp suspension (2.0 wt/v%) in ethanol for 1 hour at room temperature to adsorb the nanocrystals on the substrate. The HAp-adsorbed graft was heated at 80°C for 2 hours under vacuum (1 mm Hg) to achieve a reaction between the hydroxyl groups on the HAp nanocrystals and the alkoxysilyl groups on the poly(MPTS)-grafted polyester. The HAp nanocrystal-coated graft was washed with ethanol in an ultrasonic bath for 10 min (output: 38 kHz and 120 W) to remove unreacted HAp nanocrystals physically adsorbed on other particles. The HAp-coated graft was finally washed in a large amount of ethanol and pure water for 1 day to remove the residual organic solvents used in the polymerization process.

Compliance Measurement of the Grafts

Each two specimens of HAp-coated and uncoated grafts, each 20 mm in length, were used for the measurement of the relationship between the intraluminal pressure and external diameter. The pores of each sample were sealed with fibrin glue (Bolheal; Astellas Pharma Inc., Tokyo, Japan). One end of the graft was cannulated to an affixed stainless steel connector and the other end to a sliding connector to inject a saline solution. The latter connector was also joined to the pressure transducer (P10EZ; Gould Statham Instruments Inc., Hato Rey, Puerto Rico) to measure the intraluminal pressure. The external diameter of the graft was monitored using a CCD camera (C2400; Hamamatsu Photonics, Shizuoka, Japan) and measured with an image analyzer (C3160; Hamamatsu Photonics) at the middle portion of the specimen. The intraluminal pressure was gradually increased from 0 to 200 mm Hg (about 5 mm Hg/s). The electric outputs of a strain amplifier (6M92; NEC-Sanei, Tokyo, Japan) for the pressure transducer and the image analyzer were simultaneously sampled by an A/D converter, and the data were stored on a hard disk.

The diameter compliance, Cd, was calculated using the following formula.

Cd = ΔD/(D × ΔP), where D is the external diameter of the graft, ΔP is the pressure increment, and ΔD is the diameter change associated with ΔP; D is that at an intraluminal pressure of 100 mm Hg. The Cd was expressed in percent of diameter change per pressure increase of 0.01 mm Hg.11

Animal Experiments

The animal experiments in this study were approved by the Animal Subjects Committee of the National Cardiovascular Center. Twenty-two bilateral common carotid arteries of 11 mongrel dogs weighting 20–30 kg were used in this experiment. The animals were administered with 30 mg/kg of ampicillin sodium (Meiji Seika, Ltd., Tokyo, Japan) intravenously before anesthesia for the prevention of infection. A general anesthesia was induced with intravenous administration of 1 mg/kg of diazepam (Takeda Ltd., Osaka, Japan) and ketamin (Sankyo Lifetech Ltd., Tokyo, Japan) and maintained by inhalation of isoflurane (Merck, Ltd., Osaka, Japan).

After intravenous administration of 100 IU/kg of heparin (Mochida Pharma, Ltd., Tokyo, Japan), we applied clamps (Bulldog Serrefine; Fine Science Tools Inc., USA) to the common carotid artery, and resected it in lengths of 20, 30, or 50 mm, depending on the length of the vascular graft. The graft was anastomosed to the artery with interrupted sutures using 7-0 nylon suture (Nesco Suture, Alfresa Pharma, Ltd., Osaka, Japan). The HAp-coated graft was implanted into the left artery, and the uncoated graft of the same length was implanted into the right artery as the control. The animals received injections of ampicillin for 5 days after surgery for the prevention of infection.

The animals were sacrificed after retrieving the implanted grafts at 2, 4, or 12 weeks after surgery with above anesthesia. Heparin sodium (100 IU/kg) was administered intravenously to prevent excessive clotting prior to harvesting. After exposing the grafts, gross inspection of the grafts and surrounding tissues was performed with the naked eye, and the specimens were retrieved. Then they were rinsed with phosphate-buffered saline to remove intraluminal blood and resected longitudinally into two positions. After observation of their intraluminal surfaces, they were used for further examinations.

Histologic Examination

The resected position was fixed with 20% formaldehyde, and then thin paraffin-embedded sections were prepared for histologic examination. The sections were stained with hematoxylin and eosin (HE) stain, by immunological staining for endothelial cells with the antibody specific to the von Willebrand factor (vWf) (A0082; Dako Corp., Glostrup, Denmark), or by staining for smooth muscle cells with the antibody (M0851; Dako Corp.) specific to α-smooth muscle actin (α-SMA). Each specimen was examined histologically with a light microscope (Eclipse TE2000-U; Nikon, Japan). Furthermore, one slide in the HE stain of each specimen per animal was examined by 400× magnification and the number of inflammation cells, including macrophages, neutrophils, and lymphocytes, on the inner lumens of all grafts from proximal to distal anastomoses was distinguished morphologically and counted. Similarly, the number of giant cells was also counted.

Statistical Analysis

The difference in the number of inflammation cells and giant cells per centimeter was statistically examined by two-way analysis of variance (ANOVA) using HAp-coating presence/absence and the implantation periods. On detection of any significant effects of these factors, posthoc multiple comparisons were performed using Sheffe's test. The level of statistical significance was considered to be p < 0.05.


Figure 1 shows a SEM image of the HAp nanocrystal-coated polyester vascular graft after washing in an ultrasonic generator. The HAp nanocrystals were almost uniformly coated at certain intervals on the poly(MPTS)-graft surface without severe aggregation. The average incorporation rate of HAp on the graft surface was 41%. A mechanical test revealed that the mean diameter compliances, Cd, of the untreated grafts and HAp-coated grafts were 0.24%/0.01 mm Hg and 0.38%/0.01 mm Hg, respectively.

Figure 1.
Figure 1.:
SEM image of hydroxyapatite (HAp) nanocrystal-coated vascular graft.

The HAp nanocrystal-coated grafts were interposed into canine common carotid arteries for in vivo evaluation. In all animals, there were no complications or serious infections and no cerebral thrombosis or neurologic problems clinically. No implanted grafts were occluded at the time of harvest. Figure 2, A and C shows the typical inner lumens of the retrieved grafts at 2 weeks after implantation. Although both the HAp-coated and untreated grafts possessed a thrombus layer on the middle position of the graft, the degree of red thrombus on the HAp-coated surface was similar degree of thrombus to that of the untreated one in optimal appearance. At 12 weeks, a smooth surface was observed on each graft, as shown in Figure 2, B and D. The appearances of both inner lumens became almost the same at 12 weeks.

Figure 2.
Figure 2.:
Appearances of inner lumens of grafts at 2 and 12 weeks after implantation. A, B: Untreated grafts. C, D: Hydroxyapatite (HAp) nanocrystal-coated grafts. The grafts (A, C) were 2 cm in length and (B, D) were 3 cm in length at implantation.

In histologic findings, Figures 3–5 show the observations of the sections prepared at the transanastomotic endothelialization (TAE) regions of the HAp-coated and untreated grafts harvested after 2, 4, and 12 weeks of implantation, respectively. At these TAE regions of the grafts, the thrombus layer was not observed on the luminal surfaces. In the photographs, round or needle-like objects indicate the polyester fibers of the graft. The black arrowheads in Figures 3A and 4A show the inflammation cells. At 2 weeks after implantation, abundant inflammation cells were observed around the interface of the untreated grafts. On the other hand, HAp-coated grafts had fewer inflammatory cells (including leukocytes) than the untreated grafts (Figure 3D). Immunohistologic staining for the vWf revealed the endothelial cells, as shown in panels B and E. Immunohistologic staining for α-SMA indicated the SMCs, as shown in panels C and F. A wider area of α-SMA positive cells was observed in the HAp-coated grafts than in the untreated grafts.

Figure 3.
Figure 3.:
Light photomicrographs hematoxylin and eosin (HE) stain, Von Willebrand factor (vWf) immunostaining, and α-smooth muscle actin (α-SMA) staining of perianastomotic regions of (A–C) untreated grafts and (D–F) HAp-coated grafts at 2 weeks after implantation, respectively.
Figure 4.
Figure 4.:
Light photomicrographs hematoxylin and eosin (HE) stain, Von Willebrand factor immunostaining, and α-smooth muscle actin (α-SMA) staining of perianastomotic regions of (A–C) untreated grafts and (D–F) hydroxyapatite (HAp) nanocrystal-coated grafts at 4 weeks after implantation, respectively.
Figure 5.
Figure 5.:
Light photomicrographs hematoxylin and eosin (HE) stain, Von Willebrand factor immunostaining, and α-smooth muscle actin (α-SMA) staining of perianastomotic regions of (A–C) untreated grafts and (D–F) HAp nanocrystal-coated grafts at 12 weeks after implantation, respectively.

At 4 weeks after implantation, the inflammation response in the untreated grafts apparently continued, as shown in Figure 4A. On the other hand, the response of the HAp-coated ones still subsided (Figure 4D). The histologic sectional views of both samples at 12 weeks after implantation were similar to those at 4 weeks as shown in Figure 5.

Figure 6, A and B presents the number of inflammation cells and giant cells obtained from the HE-stained sections. ANOVA demonstrated a significant effect of the HAp coating on the untreated graft (p < 0.05) but no effect on the implantation periods (p > 0.05).

Figure 6.
Figure 6.:
A: The number of inflammation cells, including macrophages, neutrophils, and lymphocytes on inner lumens of grafts. B: The number of giant cells on inner lumens of grafts. *p < 0.05.


The HAp nanocrystals were coated on the polyester vascular graft through covalent bonding between the HAp nanocrystals and the poly(MPTS) on the polyester graft. The alkoxysilyl groups in the poly(MPTS) could be coupled with the hydroxyl groups on the HAp surfaces, which was confirmed by Fourier transform infrared spectroscopy.7 We also demonstrated the formation of covalent bonding between the HAp nanocrystals and the poly(MPTS) grafted on the substrates by measuring the adhesion strength of the HAp nanocrystals with a contact-mode atomic force microscope.12 For the graft polymerization of vinyl monomers on the polymer substrates, there are many methods. In our case of the graft polymerization of MPTS, H2O2 in benzyl alcohol was used as the initiator according to the method of Hebeish et al.,10 because of easier and better advancement, as in our previous report.4 The nanocrystals prepared using antisintering agents8,9 show almost single-crystal dispersion without the formation of secondary aggregation in a medium. Therefore, the grafts were coated the HAp nanocrystals without severe aggregation (Figure 1).

The diameter compliance, Cd, of the vascular graft is a determinant of the mechanical properties, and compliance matching for a host artery has significant importance in graft healing.11 Because a polyester graft has a lower compliance than the host arteries and even an expanded polytetrafluoroethylene (ePTFE) graft, which have the compliances of 6.8%/0.01 mm Hg and 0.51%/0.01 mm Hg, respectively,11 surface modification treatment with HAp of the polyester grafts should not decrease the compliance. It was thought that the crystallinity of the polyester fibers might be decreased in the graft polymerization process due to the use of the organic solvents and resulted in a slight increase in the compliance.

As inner lumen of optimal appearance (Figure 2), although the macroscopic views of the inner lumen of the retrieved grafts differed little, differences between the untreated and HAp-coated grafts were recognized by histologic observation at the point of early implantation. It is well known that a vicious cycle of nonhealing on Dacron polyester graft happens at an early stage because of the repetitive formation of red and white thrombi.13 In our latest data, fibroblast-like cells, which adhered on the HAp nanocrystal-coated surfaces, were hard to remove from in vitro substrates by trypsin-EDTA treatment (unpublished data). In contrast with the untreated graft, a less vicious cycle of thrombi occurrence on the HAp-coated graft might result in tight platelet adhesion on the heterogeneous surface at an early stage. The thrombi of HAp-coated and untreated grafts decreased with an increase in the implantation period.

In histologic observation (Figures 3–5), the healing in HAp-coated grafts demonstrated more progress than that in the untreated grafts at 2 and 4 weeks. Although leukocytes mass including giant cells were observed in the vicinity of both material surfaces, there were fewer on the HAp-coated surface than on the untreated one, shown with black arrowheads in Figures 3 and 4. Especially, the maturation of the vascular wall had progressed in HAp-coated grafts. The histologic finding in TAE region at 12 weeks became almost the same. At an early period of implantation, tissue healing seemed to progress better in the HAp-coated grafts than in the untreated ones. TAE is mainly limited by the proliferation of endothelial cells.13 It is also considered that endothelial precursor cells did not adhere on HAp nanocrystals of the polyester surface directly, because the substrate surfaces were immediately covered with fibrin and platelets under blood flow in vivo.

In the number of the inflammation cells and giant cells (Figure 6), not only the vascular tissue response but also the subcutaneous tissue reaction of the HAp nanocrystal-coated percutaneous devices have already been examined by implantation in the backs of Japanese white rabbits at an early stage.14 The number of inflammation cells which infiltrated the device, such as pseudoeosinophilic leukocytes and macrophages, was very low in the HAp-coated cuff at 3 and 7 days compared with the polyester cuff. In general, it is well known that HAp shows excellent adsorbability of cell-adhesive proteins, such as fibronectin and vitronectin.15 In our case, predominantly, the adsorption of own cell-adhesive proteins in the blood on HAp nanocrystals coated on the polyester vascular graft is considered to contribute to a decrease in inflammation at an early implantation stage. The results indicate that our HAp nanocrystal coating possesses good soft-tissue compatibility in the vascular system at an early implantation stage. In other words, it can be said that HAp nanocrystal coating relatively suppresses dynamic responses to untreated surface arising on the interface between biological content and an artificial material surface at an acute stage in vivo. Many persistent giant cells inhibit many more capillary and other cells ingrowth.


A novel composite artificial graft revealed that the number of inflammation cells found in the sections of HAp-coated grafts was significantly lower than that of untreated grafts in the early stage. Thus, it is demonstrated that HAp nanocrystal-coated grafts possess soft-tissue compatibility in the vascular system. The HAp nanocrystal-coating protects against inversion immunologically at an acute stage for tissue construction around vascular grafts made of artificial materials, such as polyester, polyurethane, and ePTFE. Furthermore, the improvement of cell adhesion on the graft due to the HAp coating described in our former report4 is expected to benefit previous seeding methods of endothelial cells in vitro. Noticing that recent stem cell technologies have opened the possibility for the abundant supply of endothelial cells, initial cell detachment under blood flow is one of the serious hazards for endothelialization. The HAp nanocrystal-coated grafts developed also have high possible merit for various applications, including cell seeding.


This study was supported by a Research Grant for Cardiovascular Diseases from the Ministry of Health, Labor and Welfare, Japan. The authors thank Prof. Isao Narama of the Department of Pathology, Faculty of Pharmaceutical Sciences, Setsunan University, for his kind advice on the histologic observations.


1.Maris Y, Chakfe N, Guidoin R: An albumin-coated polyester artificial graft: In vivo assessment of biocompatibility and healing characteristics. Biomaterials 17: 3–14, 1996.
2.Rumisek J, Wade C, Kaplan K, et al: Influence of early surface thromboreactivity on long-term arterial graft patency. Surgery 105: 654–661, 1989.
3.Huang-Lee LHH, Cheung DT, Nimni ME: Biochemical changes and cytotoxicity associated with the degradation of polymeric glutaraldehyde crosslinks. J Biomed Mater Res 24: 1185–1201, 1990.
4.Furuzono T, Masuda M, Okada M, et al: Increase in cell adhesiveness on a poly(ethylene terephthalate) fabric by sintered hydroxyapatite nanocrystal coating in the development of an artificial blood vessel. ASAIO J 52: 315–320, 2006.
5.Okumura M, Ohgushi H, Dohi Y, et al: Osteoblastic phenotype expression on the surface of hydroxyapatite creramics. J Biomed Mater Res 37: 122–129, 1997.
6.Tsuji T, Aoki H, Muneoka, et al: Development of vascular access made of sintered hydroxyapatite tubing. J Ceram Soci Jpn 95:792–797, 1987. (in Japanese)
7.Furuzono T, Yasuda S, Kimura T, et al: Nano-scaled hydroxyapatite/polymer composite IV. Fabrication and cell adhesion properties of a three-dimensional scaffold made of composite material with a silk fibroin substrate to develop a percutaneous device. J Artif Organs 7: 137–144, 2004.
8.Okada M, Furuzono T: Fabrication of high-dispersibility nanocrystals of calcined hydroxyapatite. J Mater Sci 41: 6134–6137, 2006.
9.Okada M, Furuzono T: Nano-sized ceramic particles of hydroxyapatite calcined with an anti-sintering agent. J Nanosci Nanotech 7: 848–851, 2007.
10.Hebeish A, Shalaby SE, Bayzeed AM: H2O2-induced graft polymerization of acrylic acid on poly(ethylene terephthalate) fibers. J Appl Poly Sci 26: 3245–3251, 1981.
11.Sonoda H, Urayama S, Takamizawa K, et al: Compliant design of artificial graft: Compliance determination by new digital X-ray imaging system-based method. J Biomed Mater Res 60: 191–195, 2002.
12.Okada M, Furukawa K, Serizawa T, et al: Interfacial interactions between calcined hydroxyapatite nanocrystals and substrates. Langmuir 25: 6300–6306, 2009.
13.Noishiki Y: Evaluation of new vascular prosthesis fabricated from ultrafine polyester fiber. Trans ASAIO J 32: 309–314, 1984.
14.Furuzono T, Ueki M, Kitamura H, et al: Histological reaction of sintered nano-hydroxyapatite-coated cuff and its fibroblast-like cell hybrid for an indwelling catheter. J Biomed Mater Res B Appl Biomater 89B: 77–85, 2009.
15.Kilpadi KL, Chang PL, Bellis SL: Hydroxyapatite binds more serum proteins, purified integrins, and osteoblast perecusor cells than titanium or steel. J Biomed Mater Res 57: 258–267, 2001.
Copyright © 2010 by the American Society for Artificial Internal Organs