Journal of Artificial Organs

, Volume 16, Issue 1, pp 59–65

Implantation study of small-caliber “biotube” vascular grafts in a rat model


    • Division of Medical Engineering and MaterialsNational Cerebral and Cardiovascular Center Research Institute
    • Department of Cardiovascular SurgeryKyoto Prefectural University of Medicine
  • Hatsue Ishibashi-Ueda
    • Department of PathologyNational Cerebral and Cardiovascular Center
  • Akihide Yamamoto
    • Department of Biomedical ImagingNational Cerebral and Cardiovascular Center Research Institute
    • Department of Medical Physics and Engineering, Division of Health Sciences, Graduate School of MedicineOsaka University
  • Hidehiro Iida
    • Department of Biomedical ImagingNational Cerebral and Cardiovascular Center Research Institute
    • Department of Medical Physics and Engineering, Division of Health Sciences, Graduate School of MedicineOsaka University
  • Taiji Watanabe
    • Division of Medical Engineering and MaterialsNational Cerebral and Cardiovascular Center Research Institute
    • Department of Cardiovascular SurgeryKyoto Prefectural University of Medicine
  • Keiichi Kanda
    • Department of Cardiovascular SurgeryKyoto Prefectural University of Medicine
  • Hitoshi Yaku
    • Department of Cardiovascular SurgeryKyoto Prefectural University of Medicine
    • Division of Medical Engineering and MaterialsNational Cerebral and Cardiovascular Center Research Institute
Original Article

DOI: 10.1007/s10047-012-0676-y

Cite this article as:
Yamanami, M., Ishibashi-Ueda, H., Yamamoto, A. et al. J Artif Organs (2013) 16: 59. doi:10.1007/s10047-012-0676-y


We developed autologous vascular grafts, called “biotubes,” by simple and safe in-body tissue architecture technology, which is a practical concept of regenerative medicine, without using special sterile conditions or complicated in vitro cell treatment processes. In this study, biotubes of extremely small caliber were first auto-implanted to rat abdominal aortas. Biotubes were prepared by placing silicone rods (outer diameter 1.5 mm, length 30 mm) used as a mold into dorsal subcutaneous pouches in rats for 4 weeks. After argatroban coating, the obtained biotubes were auto-implanted to abdominal aortas (n = 6) by end-to-end anastomosis using a custom-designed sutureless vascular connecting system under microscopic guidance. Graft status was evaluated by contrast-free time-of-flight magnetic resonance angiography (TOF-MRA). All grafts were harvested at 12 weeks after implantation. The patency rate was 66.7 % (4/6). MRA showed little stenosis and no aneurysmal dilation in all biotubes. The original biotube had wall thickness of about 56.2 ± 26.5 μm at the middle portion and mainly random and sparse collagen fibers and fibroblasts. After implantation, the wall thickness was 235.8 ± 24.8 μm. In addition, native-like vascular structure was regenerated, which included (1) a completely endothelialized luminal surface, (2) a mesh-like elastin fiber network, and (3) regular circumferential orientation of collagen fibers and α-SMA positive cells. Biotubes could be used as small-caliber vascular prostheses that greatly facilitate the healing process and exhibit excellent biocompatibility in vascular regenerative medicine.


BiotubeVascular graftsAutologous tissueIn vivo tissue engineeringConnective tissue


Small-caliber arterial substitutes are needed for cardiac and peripheral revascularization procedures. For such small artery bypass grafting procedures, autologous arterial (e.g., internal thoracic artery and radial artery) or venous (e.g., saphenous vein) grafts are still the ideal vascular substitute [13]. However, many patients do not have a vessel suitable for use owing to the poor quality, inadequate size or length, or previous harvesting of such vessels [4]. Moreover, a second surgical procedure is required in order to obtain the necessary vessel initially. Small-caliber arterial substitutes have generally proved inadequate largely because of the formation of thromboses and intimal hyperplasia [5, 6].

We developed autologous prosthetic tissues using “in-body tissue architecture” technology, which is a novel and practical approach of regenerative medicine based on the tissue-encapsulation phenomenon of foreign materials in living bodies [7]. This technology has the following advantages. The tissue prostheses can be fabricated in a wide range of shapes and sizes to suit the need of individual recipients, and most importantly, these prostheses do not require complex in vitro cell management procedures or exceptionally clean laboratory facilities, which are expensive and time consuming. This technology has been used for the development of cardiovascular tissues such as vascular grafts, namely biotubes [711], or heart valves, namely biovalves [12, 13]. Previously, the biotubes with 3-mm diameter were implanted into rabbit carotid arteries [10]. A high patency rate (9/11) was obtained at 12 weeks with endothelialization, dense collagen fibers with regular circumferential orientation, and a few elastin fibers. No aneurysm formation, rupturing, or stenosis was observed for up to 26 months without significant neointimal thickening [11].

In this study, further small-caliber biotubes with an inner diameter of 1.5 mm were auto-implanted to rat abdominal aortas for 12 weeks, and histological changes of biotubes after implantation were evaluated.

Materials and methods

Preparation of biotubes

A total of six adult female Wistar rats (age 8 weeks, weight 238.8 ± 58.6 g) were used in this experiment. All animals received care according to the Principles of Laboratory Animal Care (formulated by the National Institutes of Health, publication no.56-23, received 1985), and the research protocol (no. 9044) was approved by the ethics committee of the National Cerebral and Cardiovascular Center.

A silicone rod (diameter, 1.5 mm; length, 30 mm; Tigers Polymer, Osaka, Japan) was used as a mold. The rats were anesthetized with 1.5 % isoflurane (v/v air). A small incision was made in the shaved dorsal skin, and three molds were placed in dorsal subcutaneous pouches of each animal. After 4 weeks, the rats were anesthetized with 1.5 % isoflurane, and the implants were harvested. The three biotubes (internal diameter 1.5 mm) per one rat were obtained from the implants after trimming the peripheral tissues and pulling out the rods. One of the biotubes was used for autologous transplantation, and the others were used for measurement of burst strength or histological evaluation.

Measurement of burst strength

Biotubes (n = 6) and native abdominal aortas (n = 6) were used as samples. The native abdominal aortas were obtained at sacrificing for the implanted biotubes. One end of the biotubes was closed by tying with 4-0 silk threads and held at fixed distance to restrict the longitudinal distension and elongation. To the other end, a stainless-steel tube of 1.5 mm in external diameter was fixed with 4-0 silk threads to act as a cannula. Saline solution was pumped into the luminal cavities of the biotubes through the stainless steel tube at a rate of 50 mmHg/s until the biotubes ruptured. The burst strength was denoted by the water pressure at the instant of tube rupture, as measured by a pressure transducer (N5901; Nihon Denki Sanei, Inc., Tokyo, Japan).

Implantation of biotubes

Biotubes (length 20 mm; n = 6) were implanted to the infrarenal abdominal aorta using “the sutureless vascular connecting system” under microscopic guidance, similar to our previously reported system [14]. The connecting system consists of an introducing sheath and a connecting device. The introducing sheath was prepared by remodeling the catheter tubing (polyurethane of the 16G BD Insyte™, Becton, Dickinson and Company, NJ, USA). The connecting devices were custom-designed from the catheter (polyurethane, 16G BD Insyte™, Becton, Dickinson and Co., Franklin Lakes, NJ, USA). The dimensions of the devices were: 1.5 mm in length, 1.7 mm in external diameter, and 1.3 mm in internal diameter. In the wall of the catheter, circular micropores were processed by a CAD-assisted YAG laser ablation technique. Pore diameter was 400 μm, pore-to-pore distance was 500 μm, and the pore area to the entire area was 60 %.

Biotubes were treated by coating with argatroban (1 mg/cm2; Mitsubishi Chemical Co., Tokyo, Japan) for 10 min in order to make it antithrombogenic immediately before implantation. The connecting devices were inserted into the proximal and distal ends of the rat aorta through the introducing sheath. Subsequently, the treated aortic ends were inserted into biotubes, followed by banding from the outside of the biotubes with 7-0 silk threads (Alfresa, Osaka, Japan). Patency was examined at the time of surgery by direct inspection. The wound was closed with 4-0 nylon sutures. Thereafter, the rats had free access to standard food and water. No anticoagulants or antiplatelet agents were administered postoperatively. After 12 weeks of implantation all biotubes were harvested.

Evaluation of graft status

Graft status was evaluated at 6 and 12 weeks after transplantation by 3-T contrast-free time-of-flight magnetic resonance angiography (TOF-MRA) under anesthesia induced by an intramuscular injection of pentobarbital (40 mg/kg). A human whole-body 3-T magnetic resonance imaging (MRI) scanner (Sigma, GE Healthcare, Milwaukee, WI, USA) was employed. The gradient coil system was capable of providing a maximum gradient amplitude of 40 mT/m. All sequence programs employed in this study were designed for clinical studies. A developed single-turn surface coil of 62 mm diameter was used for MR imaging. Contrast-free TOF-MRA was performed using a three-dimensional flow-compensated fast spoiled gradient recalled (3D-FSPGR) sequence [repetition time (TR) = 21 ms, echo time (TE) = 5.4 ms (out of phase), flip angle (FA) = 15°, slice thickness = 0.4 mm, field of view (FOV) = 80 mm × 60 mm, matrix = 288 × 192, locs per slab = 128, the number of excitations (NEX) = 1, scanning time = 5 min 58 s]. For suppressing venous signals, a region of 40 mm thickness on the caudal side of the measured slab was saturated. MR angiograms were analyzed by generating the partial maximum intensity projection (pMIP) with a commercial software package (AZE, Tokyo, Japan).

Histological examination

The specimens of the implanted biotubes were fixed with 10 % formalin, embedded in paraffin, sliced into short axis cross sections, and finally stained with hematoxylin-eosin, Masson’s trichrome, Elastica van Gieson, or Siriusu red. Immunohistochemistry was performed using monoclonal antibodies against α-smooth muscle actin (anti-human α-SMA mouse monoclonal antibody clone 1A4; Dako Japan, Kyoto, Japan; 1:100 dilution) and factor VIII (anti-human von Willebrand factor mouse monoclonal antibody clone F8/86; Dako Japan, Kyoto, Japan; 1:100 dilution).


Preparation of biotubes

After the molds (Fig. 1a) were inserted into the dorsal subcutaneous pouches of rats for 4 weeks, all molds were completely encapsulated with very thin homogeneous autologous connective tissues (Fig. 1b). Biotubes were obtained by removing the molds from capsule tissues (Fig. 1c). The efficiency of the biotube preparation was 100 %. The burst pressure of biotubes was 1085 ± 525 mmHg (mean ± SD), which is approximately half of that of rat aorta (2478 ± 628 mmHg).
Fig. 1

a Silicone rods (diameter 1.5 mm, length 15 mm) were used as the molds. b The molds were embedded into dorsal subcutaneous pouches of rats for 4 weeks. The molds were completely encapsulated with connective tissues. c The obtained biotubes with thin wall (thickness 56.2 ± 26.5 μm). Scalebar 1.0 mm

Biotube implantation

Biotubes could be implanted easily into the rats using a custom-designed sutureless vascular connecting system, where no suturing treatment for anastomosis was needed (Fig. 2a). Total ischemia time for implantation was very short (34.8 ± 12.1 min). No bleeding was observed after declamping (Fig. 2b).
Fig. 2

a Insertion of two connecting devices, fabricated from catheter tubing (polyurethane of the 16G BD InsyteTM, Becton, Dickinson and Company, Franklin Lakes, NJ, USA), into the proximal and distal ends of the rat abdominal aorta through the introducing sheath under clamping. b The treated aortic ends were inserted into the biotube vascular graft, followed by banding from the outside of the biotube with 7-0 silk threads. Pulsation of the biotubes was noted immediately after declamping. c After 12 weeks of implantation, the biotube had little adhesion with surrounding tissues. d Macroscopic observation revealed an extremely flat luminal surface including the luminal regions of the connecting devices, which were completely impregnated into the vascular tissues (scale bar 3 mm)

MRA distinctly visualized the patent graft connected to the abdominal aorta together with renal arteries and common iliac arteries (Fig. 3). However, pseudostenosis by artifact of the connection was observed in both anastomosis regions (arrows in Fig. 3). In the midgraft region, little stenosis and no aneurysmal dilation were observed in all patent grafts. The overall patency rate was 66.7 % (4/6). Two of the six grafts were occluded by thrombosis at 6 weeks after implantation.
Fig. 3

MRA images of the rat abdominal aorta at 1 (a), 6 (b) and 12 (c) weeks after implantation of biotubes. Upper side of all photos indicates the proximal side. White arrows indicate the proximal and distal anastomosis regions of the abdominal aorta. A mechanical stenotic lesion, which may have been due to the anastomosis, was observed in both anastomosis regions. In the midgraft region, neither significant stenosis nor aneurysmal dilation was observed

Implanted biotubes were easily harvested with no damage because there were few adhesions between biotubes and surrounding tissues (Fig. 2c). In the obtained biotubes 12 weeks after implantation, macroscopic observation revealed an extremely flat surface including the region of connecting device, which was completely impregnated into the developed thin tissues (Fig. 2d). In the circumferential cross section of midgraft region, the wall had grown thick, keeping the area of the lumen. Before implantation, wall thickness of the biotubes was 56.2 ± 26.5 μm (Fig. 4a-1). At 12 weeks after implantation, neointima was formed with wall thickness of 235.8 ± 24.8 μm (Fig. 4b-1).
Fig. 4

The circumferential cross section of the biotube before implantation (a-16) and obtained at 12 weeks after implantation (b-16). The circumferential cross section of native rat abdominal aorta (c-13). Stained with H&E (a-1, b-1, c-1), Masson trichrome (a-2, b-2, c-2), Elastica van Giesson (a-3, b-3, c-3), factor VIII (a-4, b-4), α-SMA (a-5, b-5), and Sirius red (a-6, b-6). Masson trichrome stain revealed that it was composed of collagen-rich tissue. Polarizing microscopic observation after Sirius red stain showed circumferential orientation of the collagen fiber after implantation (b-6). Elastica van Giesson staining revealed an elastic fiber network in the neointima of the biotube after implantation (b-3). Immunohistological staining for factor VIII and α-smooth muscle actin revealed α-smooth muscle actin-positive cells (b-5) and endothelial lining at the luminal surfaces (b-4) of biotube after implantation (scale bar 100 μm)

Histological change

Before implantation biotubes were composed of collagen-rich tissue with no elastic fibers (Fig. 4a-2, a-3) and collagen fibers were randomly attached (Fig. 4a-6). No abnormal collection or infiltration of inflammatory cells was observed. The main cell component was fibroblasts with no vascular constituent tissues (Fig. 4a-4, a-5).

On the other hand, after 12 weeks of implantation the neointima was segregated in two layers, where the luminal layer with a mesh-like elastin fiber network (Fig. 4b-3) and circumferential oriented collagen fiber (Fig. 4b-2, b-6) were formed on the almost cell-free dense collagen basement layer. The construction of collagen and elastin fibers was very similar to that of native rat abdominal aorta (Fig. 4c-1, c-2). The luminal layer had many α-SMA-positive, myofibroblasts or smooth muscle cells (Fig. 4b-5) and was completely covered with endothelial cells (Fig. 4b-4).


Tissue engineering combines the principles of engineering and biological sciences to develop viable structures that can replace diseased deficient natural structures. Some investigators have successfully implanted in vitro tissue-engineered vascular grafts in animals by using either decellularized natural tissues or biodegradable synthetic polymers as scaffolds [1518]. However, these procedures require complicated cell-management protocols, including harvesting, seeding on appropriate scaffolds, and development of neotissues by culturing cells in bioreactors under strictly sterile conditions; all of these procedures are time consuming and expensive. On the other hand, in vivo tissue engineering can produce completely autologous tissues without any artificial support materials. Campbell et al. developed autologous vascular grafts in the abdominal cavities of rats or rabbits. Although the grafts had a relatively large internal diameter (3 mm) for small-caliber native aorta of rats, the high patency rate of 67 % was reported at 4 months of implantation by layering mesothelium cells on the luminal surface of myofibroblast-walled tubes [19].

In this study, we evaluated the potential of biotubes to generate a vascular graft for extremely small arteries (internal diameter 1.5 mm). The biotubes were implanted to the abdominal aorta of rats (internal diameter 1.3 mm) by end-to-end anastomosis using custom-designed sutureless connecting devices [15]. It must be verified whether biotubes can be sutured and have sufficient burst strength to withstand physiologic blood pressure or not. Because the biotubes have adequate mechanical properties for a vascular prosthesis, they were recently applied to carotid arteries [20] and abdominal aortas of beagle dogs [21]. In all implantations for systemic circulation, including those described in this study, neither rupture nor aneurysm formation was observed.

The patency rate of the biotubes at 12 weeks after implantation was high in this study (66.7 %). Because the main components of the original biotubes were collagen fibers and fibroblasts, it was considered that acellular luminal surfaces without endothelial cell coverage carry a substantial risk for thrombosis when exposed to the blood directly. In our previous implantation study of biotubes, all grafts without anti-thrombogenic coating were completely occluded at 2 weeks after implantation [10]. Strong anti-thrombogenicity, provided by the complete endothelialization of the luminal surfaces, is highly desirable. For this reason, most recent studies have focused on the creation of tissue-engineered cardiovascular implants using autologous endothelial cell seeding and bioreactor culturing prior to implantation. Seeding with autologous vascular cells on the luminal surface has provided a much higher patency rate than non-cell-seeded grafts [22]. However, such cell management and processing are complicated and invasive, and they will render the implants prone to infection. In this study, acute thrombus formation was considerably prevented by argatroban coating.

At 12 weeks after implantation, almost complete vascular tissue was re-constructed. Endothelial cells and SMCs migrated into the biotube and became organized into an endothelium and a media-like smooth muscle layer after implantation. In addition, mesh-like elastin network was formed on the basement dense collagen layer. Native-like vascular structure was regenerated in a short period of implantation. Therefore, the autologous connective tissue could serve as an ideal scaffold for vascular wall formation. Some investigators have discussed which cells and extracellular matrix remodeled tissue-engineered vascular grafts after implantation. Kuwabara et al. [23] suspected endothelial cells and smooth muscle cells were regenerated from circulating blood stem cells. Hibino et al. [24] reported neovessel formation arises from ingrowth of vascular cells from the neighboring blood vessel. Erman et al. described that endothelial coverage of the luminal surface, transmural cellular infiltration, and formation of neocapillaries in the graft body are the major graft healing characteristics [17]. In this study, it still remains unclear which cells and extracellular matrix remodeled biotubes after implantation. It is necessary to investigate this issue in the future.

Since the observation by MRA is simple and non-invasive, assessment of the status of small-caliber vascular grafts could be performed in the same rat at different times. The repeatable MRA observation in a single rat enabled correct assessment of the graft status over the follow-up period. Such repeatability will reduce the variation in results stemming from individual differences in experimental animals [25]. By using the connecting system, aortic clump time was only 30 min, which is about half that in the traditional suturing method [14]. Since no bleeding at the connecting parts occurred, hemostasis was very easy in all implantations. In addition, although pseudostenosis by artifacts of the connecting devices was observed at both anastomosis regions, there was no actual stenosis. The stenosis by suturing could lead to serious occlusion in small-caliber grafts. The control of the bleeding and minimal anastomotic stenosis using the connecting system induced reproductive results in the implantation study independent of the technique of surgeons. On the other hand, there is one notable limitation to this system in that the connecting device was directly exposed to blood flow, which might promote thrombus formation at the luminal surface of the device. Because of the small caliber of the grafts, even small thrombi could result in occlusion. However, the early endothelialization promoted by the microporous mesh structure could prevent this complication. Therefore, this system minimizing the blood flow interruption to obtain complete and reliable anastomosis was appropriate for this study.


The biotube vascular grafts provided high patency even when implanted in extremely small-caliber vessels with internal diameter of 1.5 mm. An almost complete artery-like structure with cellular components of endothelial cells and smooth muscle cells and extracellular components of collagen and elastin was formed only 12 weeks after implantation. Biotubes could thus be used as small-caliber vascular prostheses that greatly facilitated the healing process and exhibited excellent biocompatibility.


The authors thank Ms. Manami Sone for her technical support in this study. This study was funded in part by a Grant-in-Aid for Scientific Research (B23360374) from the Ministry of Education, Culture, Sports, Science and Technology of Japan.

Copyright information

© The Japanese Society for Artificial Organs 2012