A rat arteriovenous graft model using decellularized vein

Abstract

Background

The high rate of clinical failure of prosthetic arteriovenous grafts continues to suggest the need for novel tissue-engineered vascular grafts. We tested the hypothesis that the decellularized rat jugular vein could be successfully used as a conduit and that it would support reendothelialization as well as adaptation to the arterial environment.

Materials and methods

Autologous (control) or heterologous decellularized jugular vein (1 cm length, 1 mm diameter) was sewn between the inferior vena cava and aorta as an arteriovenous graft in Wistar rats. Rats were sacrificed on postoperative day 21 for examination.

Results

All rats survived, and grafts had 100% patency in both the control and decellularized groups. Both control and decellularized jugular vein grafts showed similar rates of reendothelialization, smooth muscle cell deposition, macrophage infiltration, and cell turnover. The outflow veins distal to the grafts showed similar adaptation to the arteriovenous flow. Both CD34, CD90 and nestin positive cells, as well as M1-type and M2-type macrophages accumulated around the graft.

Conclusions

This model shows that decellularized vein can be successfully used as an arteriovenous graft between the rat aorta and the inferior vena cava. Several types of cells, including progenitor cells and macrophages, are present in the host response to these grafts in this model. This model can be used to test the application of arteriovenous grafts before conducting large animal experiments.

Keywords

Arteriovenous graft decellularization macrophage CD90 endothelial progenitor cell smooth muscle cell

Introduction

The native arteriovenous fistula (AVF) is currently the best available conduit for dialysis access, but AVF are still associated with a very high failure rate.¹ AVF require a vein to be available for both direct anastomosis to the artery as well as for outflow vessel cannulation; however, due to insufficient venous length or sclerosis secondary to venipuncture, veins are not always available and arteriovenous grafts (AVG) are frequently required.¹

Prosthetic grafts for dialysis access are still associated with a lower rate of primary patency and a higher rate of infection compared to AVF.² Commonly used prosthetic conduits such as Dacron and expanded polytetrafluoroethylene show relatively high rates of thrombosis and infection, leading to several described modifications that attempt to reduce these complications;³ however, results are frequently not as clinically useful.⁴ Tissue engineering to create novel vascular grafts is an emerging strategy with great promise for clinical translation.^5,6 Decellularized vessels have similar extracellular matrix and similar structural architecture as native vessels, but with reduced immunogenicity.⁷ Several studies have showed encouraging results of decellularized vessels in animal studies,^8,9 and there are recent human clinical trials of bioengineered human acellular vessels for dialysis access.¹⁰

We have developed a rodent AVG model by using decellularized heterologous (from a different animal) rat carotid artery (CA) that shows excellent patency in a small animal model, similar to control autologous (from the same animal) CA grafts.¹¹ However, venous conduits are generally preferred and more commonly used for dialysis access, and whether decellularized veins can function as an arteriovenous conduit similar to native veins is not currently known. Accordingly, we hypothesized that decellularized jugular vein (JV) can be used as an AVG conduit successfully in this rodent model and that decellularized veins do not undergo accelerated neointimal hyperplasia.

Methods

Animal model

All experiments were approved by the Institutional Animal Care and Use Committee at Zhengzhou University; all experiments were also carried out in accordance with the NIH guidelines for the care and use of laboratory animals (NIH Publication #85–23 Rev. 1985). Male Wistar rats (6–8 weeks) were used as previously described.¹¹ Briefly, anesthesia was given using 10% chloral hydrate (0.2–0.3 mL/100 g IP). After anesthesia was given and determined to be adequate, a midline incision was made in the neck, and the right JV branch was exposed; an approximately 1 cm length of the vein (diameter 1 mm) was harvested, and then the neck incision was closed. The JV was temporarily stored in saline on ice (approximately 10 min) prior to implantation. A midline abdominal incision was then made, and the infrarenal inferior vena cava (IVC) and the aorta were exposed. One end of the control (n = 6) or decellularized JV (n = 6) was sewn to the IVC and the other to the aorta in end-to-side fashion using running 10–0 nylon sutures, with care given to avoid any tension in the graft; the graft was oriented perpendicularly to the IVC and aorta, and blood flow in the vein was antegrade. After completion the anastomoses, the clamps were removed and hemostasis assured. The abdomen was then closed, and the rat was allowed to recover from anesthesia.

We have previously demonstrated that mouse AVF mature with a stable amount of wall thickening by day 21 in an aortocaval model;¹² similarly, in the rat CA AVG model, the AVG grafts, aorta, and IVC were harvested at day 21 to assess maturation.¹¹ Based on these results, the samples in this model were examined on postoperative day 21. The patency of the AVG was directly observed by the rapid arterial flow in the graft and outflow IVC. The aorta, IVC, and graft were explanted for analysis as described below. No immunosuppressive agents, antiplatelet agents, antibiotics, or heparin were given at any time.

A power calculation was performed to determine the number of mice in each group. To compare groups at a significance level of 0.05 and a power of 80% to detect a difference of 30% thickness of the neointimal hyperplasia, we estimated that six mice were required in each group.

Decellularization of rat JV

The right JV (1 cm length) was dissected and carefully removed using sterile technique as described above; the JV were then stored at 4°C in phosphate-buffered saline (PBS) containing penicillin 100 U/mL and streptomycin 100 µg/mL. Decellularization of JV was accomplished as described previously.⁷ Briefly, JV were incubated in 250 mL CHAPS buffer (8 mM CHAPS, 1 M NaCl, and 25 mM EDTA in PBS) for 12 h, followed by a 60 min wash, and then incubated in 10 mL sodium dodecyl sulfate buffer (1.8 mM sodium dodecyl sulfate, 1 M NaCl, and 25 mM EDTA in PBS) for 24 h, followed by another 24 h wash with PBS to completely remove the detergent. Decellularized JV were stored at 4°C until use in heterologous fashion.

Tissue analysis

Rats were anesthetized as previously described, and tissues were fixed by transcardial perfusion of PBS followed by 10% formalin. The samples were fixed overnight in 10% formalin followed by a 24-h immersion in 70% alcohol. Tissue was then embedded in paraffin and sectioned (4 µm thickness). Tissue sections were deparaffinized and stained using an EVG staining kit (Baso, Zhuhai, China) according to the manufacturer’s recommendations.

Immunohistochemistry

Sections were heated in citric acid buffer (pH 6.0, Beyotime, Shanghai, China) at 100°C for 10 min for antigen retrieval. Sections were then treated with 0.3% hydrogen for 30 min and blocked with blocking buffer (Beyotime, Shanghai, China). Sections were then incubated overnight at 4°C with primary antibodies diluted in dilution buffer (Beyotime, Shanghai, China). After overnight incubation, the sections were incubated with appropriate secondary antibodies for 1 h at room temperature and treated with DAB Horseradish Peroxidase Color Development Kit (Beyotime, Shanghai, China) to detect the reaction products. Finally, the sections were counterstained with Hematoxylin (Baso, Zhuhai, China). No primary antibody but PBS was used as negative control. High power photographs were taken, and positive cell numbers were counted and blinded reviewed by three professional pathologists; the density of the IHC that was positively stained was measured by image J software (NIH).

Immunofluorescence

Sections were heated in citric acid buffer (pH 6.0) at 100°C for 10 min for antigen retrieval. Sections were then blocked with blocking buffer (Beyotime, Shanghai, China) and then incubated overnight at 4°C with primary antibodies diluted in dilution buffer (Beyotime, Shanghai, China), The sections were incubated with secondary antibodies for 1 h at room temperature, after which sections were stained with the fluorescent dye 40,6-diamidino-2-phenylindole (Solarbio, Beijing, China) to mark cellular nuclei. High power photographs were taken, and positive cell numbers were counted and blinded reviewed by three professional pathologists.

Primary and secondary antibodies

Primary antibodies included: anti-α-actin (Abcam, ab5694); anti-cleaved Caspase-3 (Cell Signaling #9661); anti-CD31 (Abcam, ab28364); anti-CD34 (R&D, AF4117); anti-CD68 (Abcam, ab31630); anti-VEGFR2 (Abcam, ab2349); anti-nestin (Abcam, AB11306); anti-CD90 (Abcam, AB225); anti-IL-10 (Abcam, AB9969); anti-TGM2 (Abcam, AB421). Secondary antibodies used for IF were: donkey anti-goat Alexa-Fluor-488, donkey anti-rabbit Alexa-Fluor-488, and donkey anti-mouse Alexa-Fluor-568-conjugated antibodies from Invitrogen (1:1000).

Statistical analysis

Data are expressed as the mean ± SEM. Statistical significance for these analyses was determined by ANOVA and t-test. p-Values less than 0.05 were considered significant. Data were analyzed using Prism 6.0 software, and power calculations were performed using the StatMate module (GraphPad Software; La Jolla, CA).

Results

JVs were harvested from rats and decellularized; after decellularization, H&E staining showed no residual nuclei in the vein (Figure 1(a) and (b)); in particular, CD31-positive endothelial cells and α-actin-positive smooth muscle cells were not detectable using either immunohistochemistry or immunofluorescence (Figure 1(a), (c) and (d)). Both native and decellularized JV showed almost no elastin fiber staining and similar collagen-1 density (Figure 1(a) and (e)).

Figure 1.

JV before and after decellularization: (a) L, lumen; scale bar, 100 µm; n = 3; (b) number of nuclei in the control and decellularized JV, *p = 0.0556, n = 3; (c) percentage of confluent CD31-positive cells, *p < 0.0001, n = 3; (d) α-actin positive cell density, *p = 0.0010, n = 3; (e) collagen-1 density in the adventitia, *p = 0.3902, n = 3.

Decellularized JV were implanted into rats (n = 6) as heterologous AVG; autologous JV (n = 6) were implanted immediately after harvest as a control group, similar to the commonly used autologous vein (Figure 2(a)). Immediately after implantation, AVG dilated with arterial blood flow. After three weeks, AVG were patent as determined by direct inspection and were well incorporated onto the anterior walls of the aorta and IVC, with no aneurysm or thrombus formation; cross sections confirmed AVG patency from the aortic to the IVC anastomoses (Figure 2(b)). No rats died or had any evidence of thrombosis; there was a 100% technical success rate. Both the arterial anastomosis and the venous anastomosis had similar luminal areas (Figure 3(a) and (b)). There was a very thin neointima in the body of the control and decellularized JV grafts as well as in the outflow vein (Figure 3(a)), similar to that previously reported in decellularized arterial grafts.¹¹

Figure 2.

Schema of the JV arteriovenous graft and analysis: (a) diagram and surgical photographs. Scale bar, 1 mm. A, aorta; IVC, inferior vena cava; AVG, JV graft; (b) diagram showing continual cross-sectional locations. Scale bar: 1 mm or 100 µm.

Figure 3.

Control and decellularized JV grafts show similar adaptation to arterial flow: (a) elastin staining, day 21; scale bar, 1 mm; A, aorta; IVC, inferior vena cava; AVG, arteriovenous graft; n = 6; (b) area of the aortic (p = 0.8749) and IVC anastomosis (p = 0.8920); AVG neointimal thickness (p = 0.8499) and outflow IVC neointimal thickness (p = 0.8963); n = 6.

At day 21 after implantation, there was a monolayer of CD31-positive endothelial cells on the luminal side of the AVG with a similar degree of confluence between native and decellularized JV (Figure 4(a) and (b)). Immunofluorescence showed that these endothelial cells were immunoreactive with both CD34 and VEGFR2, that is identity consistent with endothelial progenitor cells (EPCs), with a similar dual-positive cell number between native and decellularized JV (Figure 4(a) and 4(b)). Similarly, luminal cells were also immunoreactive for both vWF and vascular cell adhesion molecule 1 (VCAM-1), with no difference in the numbers of dual-positive cells (Figure 4(a) and (b)). Below the luminal surface of the AVGs, there were α-actin positive cells, with similar density between native and decellularized JV (Figure 4(c) and (d)). There were also similar numbers of CD68-positive cells in the control and decellularized JV grafts (Figure 4(c) and (d)). There were also similar numbers of PCNA-positive cells, PCNA and α-actin dual-positive cells, and cleaved caspase-3-positive cells in the control and decellularized JV grafts (Figure 4(c) and (d)).

Figure 4.

Recellularization of control and decellularized JV grafts after implantation: (a) AVG, JV graft. L, AVG lumen; yellow arrow showing dual-positive cells; scale bar, 100 µm; n = 3; (b) CD31-positive cell confluence (p = 0.3415, n = 3), CD34 and VEGFR2 dual-positive cells (p = 0.6213, n = 3), vWF and VCAM-1 dual-positive cells (p = 0.6675, n = 3); (c) day 21; scale bar, 100 µm. n = 6; (d) α-actin-positive cell density (p = 0.8600, n = 3), CD68-positive cell number (p = 0.9568, n = 3), PCNA-positive cell number (p = 0.6584, n = 3), PCNA and α-actin dual-positive cells (p = 0.6213, n = 3), and cleaved caspase-3-positive cell number (p = 0.8149, n = 3).

We then examined the venous outflow distal to the venous anastomosis of the AVG (Figure 5(a)). Similar to the luminal surface of the AVG, there were CD31-positive endothelial cells with a similar confluence (Figure 5(a) and (b)) as well as CD34 and VEGFR2 dual-positive cells with a similar dual-positive cell number (Figure 5(a) and (b)); there were also vWF and VCAM-1 dual-positive cells with no difference of the dual-positive cell number (Figure 5(a) and (b)). α-Actin positive cells were below the outflow luminal surface with a similar density (Figure 5(c) and (d)). There were also similar numbers of CD68-positive cells, PCNA-positive cells, PCNA and α-actin dual-positive cells, and cleaved caspase-3-positive cells in the outflow veins (Figure 5(c) and (d)).

Figure 5.

Characterization of the outflow vein distal to the control and decellularized JV grafts: (a) L, IVC lumen; yellow arrow showing dual-positive cells; scale bar, 100 µm; n = 3; (b) CD31-positive cell confluence (p = 0.6330, n = 3), CD34 and VEGFR2 dual-positive cells (p = 0.7850, n = 3), vWF and VCAM-1 dual-positive cells (p = 0.6130, n = 3); (c) day 21; scale bar, 100 µm, n = 3; (d) α-actin-positive cell density (p = 0.9244, n = 3), CD68-positive cell number (p = 0.7318, n = 3), PCNA-positive cells (p = 0.8813, n = 3), PCNA and α-actin dual positive cells (p = 0.5879, n = 3), and cleaved caspase-3-cells (p = 0.6638, n = 3).

We also examined the peri-graft area below the graft and between the aorta and the IVC (Figure 6(a)). This area contained cells that expressed the progenitor cell markers CD34 (hematopoietic stem cells), CD90 (hematopoietic stem cells), and nestin (neuroectodermal stem cells); some cells were dual-positive for CD34 and CD90 or CD34 and nestin, but there was no significant difference in the number of CD90-positive cells or nestin-positive cells between the control and decellularized groups (Figure 6(b) and (d)). There were also some cells were dual-positive for CD68 and TGM2, consistent with M2-type macrophages, and some cells were dual-positive for CD68 and iNOS, consistent with M1-type macrophages; there was no significant differences in the number of CD68 and TGM2 dual-positive cells or CD68 and iNOS dual-positive cells between the control and decellularized groups (Figure 6(c) and (e)).

Figure 6.

CD90 and nestin-positive cells and M1 and M2 macrophages accumulate around the arteriovenous graft: (a) day 21, dotted circle indicates area of interest; scale bar, 1 mm; A, aorta; IVC, inferior vena cava; (b) yellow arrow shows positive cells; n = 3; (c) yellow arrow shows dual-positive cells; n = 3; (d) number of CD90-positive cells (p = 0.6422, n = 3) and nestin-positive cells (p = 0.7607, n = 3); (e) number of CD68 and TGM2 dual-positive cells (p = 0.8541, n = 3) and CD68 and iNOS dual-positive cells (p = 0.6433, n = 3).

Discussion

This study shows that decellularized JV implanted as a heterologous AVG in rats have excellent patency with minimal thickening or dilation (Figure 3); this may be due to repopulation with several vascular cell types, including endothelial cells, smooth muscle cells, and macrophages (Figures 4 and 5) and several stem cell types (Figure 6) as well as the high blood flow from the aorta to IVC through the graft. These data suggest that decellularized veins may be a potential source for AVG, similar to our previous data showing the utility of the decellularized artery.¹¹

Our finding that decellularized veins heal by host cell infiltration and proliferation as well as neointimal reendothelialization is similar to our previous findings to the healing processes that occur in acellular pericardial or polyester patches.^13–15 We have previously shown that decellularized artery also shows luminal neointimal reendothelization and infiltration of several host cell types in this rat AVG model; thus, the adaptation of both CA and JV grafts is similar, with both conduits characterized by a high rate of success with no thrombosis formation; we did not find any specific differences between our data reported here and that using a CA graft in the same model.¹¹ Other groups have also reported a similarly high rate of success using AVG in rats.¹⁶,¹⁷ We believe that the use of decellularized vein as a conduit has been studied infrequently since the use of a prosthetic conduit in a human patient implies limited availability of autologous vein (vein from the patient); however, use of a decellularized vein in heterologous fashion may increase the translatable potential of this work.

There is still debate regarding the origin of the neointimal cells, with speculation that these cells derive from local resident cells in the venous limb of the fistula or the arterial anastomosis, but not from the bone marrow;^18,19 however, bone marrow-derived EPC, derived from the bone marrow, may also play a role in neointimal formation.^20,21 We show smooth muscle cells, EPC (Figures 4 and 5), macrophages, CD90, and nestin-positive cells in the wall of both of the AVG and the outflow IVC (Figure 6), suggesting that both the bone marrow as well as local cells from the adjacent anastomoses may play a role in cell deposition and neointimal reendothelialization. Interestingly, the finding of CD90- and nestin-positive cells (Figure 6) is similar to the finding of these cells in explanted bioengineered human acellular vessels²² and suggests that these progenitor cells may play a role in the transformation of a decellularized graft into a functional living graft. This rat AVG model can be a useful tool to examine acellular AVGs; however, a limitation of this study is that our data are based on healthy rats not rats with chronic renal failure, limiting directly clinical applicability. Although some decellularized grafts can show aneurysmal degeneration,²³,²⁴ we did not observe any aneurysmal degeneration in this model at three weeks. We did observe that two rats died after three weeks secondary to heart failure that was likely induced by the high aortocaval flow in this model.

Neointimal formation after vascular graft implantation is a complex process, with many factors still not well understood, especially in the venous wall.^25–28 Multiple factors have been implicated in neointimal formation but require additional mechanistic understanding, including fibrin, platelet, and plasma adhesion to the graft surface; the temporal sequence of macrophage, leukocyte, and lymphocyte migration and proliferation; and smooth muscle cell proliferation and endothelial cell reendothelialization. Despite some understanding of the components that form neointimal hyperplasia, the clinical problem has not yet been fully solved. Since there is still no commonly available rodent AVG model, other than our previously reported decellularized CA model,¹¹ this is the first animal model to report the use of a potentially clinically available conduit, that is a decellularized vein. However, this report did not focus on repeat cannulation of the graft but focused on neointimal hyperplasia; we also did not measure the mechanical properties of the conduits. In addition, this model did not include renal failure conditions, and thus, there may be some differences in this model when implanted into rats with renal failure. In this AVG model, we did not notice any thrombus formation in the anastomosis or the outflow; this is similar to many other AVG or AVF models that also did not report any thrombosis.^12,29–32 Lack of thrombosis in this model is not surprising since the aorta and IVC are directly connected together, forming a high flow aortocaval shunt; in addition, we did not observe thrombosis formation at this early timepoint. Although it is possible that thrombus may form at later times, the heart failure that forms in response to the aortocaval shunt may limit survival, thrombus formation, and neointimal hyperplasia.

Although there is a large body of work relating to decellularized conduits including studies in animal models^17,33–35 and humans,³⁶,³⁷ these models have typically been performed using large animals. This rodent AVG model is an easily performed small animal model; however, it does not fully mimic human conditions since it is not superficial but in the abdomen, and it cannot be punctured because of the small diameter. In addition, rodent models do not develop accelerated neointimal hyperplasia as some large animals such as pigs.³⁸ Furthermore, the applicability of the decellularizing agents (CHAPS and sodium dodecyl sulfate buffers) to human studies is not established, and it is not yet known what the long-term outcomes are, especially potential to show aneurysmal degeneration. However, this model is a novel tool to explore the failure of AVG grafts with lower costs and increased convenience compared to large animal models.

In summary, decellularized JV implanted as a heterologous AVG shows similar remodeling as control autologous JV in this model, with infiltration of macrophages and progenitor cells that contribute to excellent patency. This rodent model is a useful tool to test novel AVG material prior to large animal experiments.

Footnotes

Authors’ contribution

HB and AD designed experiments, performed data analysis, wrote, and revised the manuscript; HB, WW, WL, and SW conducted animal models, histological experiments, and compiled data; ZW and ML compiled data; HB, ZW, YX, and AD obtained funding.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical statement

The study was approved by the Institutional Animal Care and Use Committee.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This study was funded by the National Natural Science Foundation of China (Grant number 81870369).

ORCID iD

Hualong Bai

References

Patel

Hanisch

, et al. Future research directions to improve fistula maturation and reduce access failure. Semin Vasc Surg 2016; 29: 153–171.

Jack

Cronenwett

JK.

Rutherford’s vascular surgery. 8th ed. Philadelphia, PA: Elsevier; 2014, 1334p.

Glickman

Drug eluting grafts for hemodialysis access. J Vasc Access 2017; 18: 53–55.

Zea

Menard

, et al. Heparin-bonded polytetrafluorethylene does not improve hemodialysis arteriovenous graft function. Ann Vasc Surg 2016; 30: 28–33.

Gage

Lawson

JH.

Bioengineered hemodialysis access grafts. J Vasc Access 2017; 18: 56–63.

Benrashid

McCoy

Youngwirth

, et al. Tissue engineered vascular grafts: origins, development, and current strategies for clinical application. Methods 2016; 99: 13–19.

Gui

Muto

Chan

, et al. Development of decellularized human umbilical arteries as small-diameter vascular grafts. Tissue Eng Part A 2009; 15: 2665–2676.

Quint

Kondo

Manson

, et al. Decellularized tissue-engineered blood vessel as an arterial conduit. Proc Natl Acad Sci USA 2011; 108: 9214–9219.

Dahl

Koh

Prabhakar

, et al. Decellularized native and engineered arterial scaffolds for transplantation. Cell Transpl 2003; 12: 659–666.

10.

Lawson

Glickman

Ilzecki

, et al. Bioengineered human acellular vessels for dialysis access in patients with end-stage renal disease: two phase 2 single-arm trials. Lancet 2016; 387: 2026–2034.

11.

Bai

Dardik

Xing

Decellularized carotid artery functions as an arteriovenous graft. J Surg Res 2019; 234: 33–39.

12.

Yamamoto

Protack

Tsuneki

, et al. The mouse aortocaval fistula recapitulates human arteriovenous fistula maturation. Am J Physiol Heart Circ Physiol 2013; 305: H1718–H1725.

13.

Bai

Lee

, et al. Transforming growth factor-beta1 inhibits pseudoaneurysm formation after aortic patch angioplasty. Arterioscler Thromb Vasc Biol 2018; 38: 195–205.

14.

Kuwahara

Hashimoto

Tsuneki

, et al. CD44 promotes inflammation and extracellular matrix production during arteriovenous fistula maturation. Arterioscler Thromb Vasc Biol 2017; 37: 1147–1156.

15.

Bai

Wang

Foster

, et al. Pericardial patch venoplasty heals via attraction of venous progenitor cells. Physiol Rep 2016; 4: pii: e12841.

16.

Cuadros

Hughes

JE.

Patency of synthetic microvenous conduits: an experimental study in the femoral vein of the rat. Plast Reconstr Surg 1986; 78: 378–382.

17.

Kokozidou

Katsargyris

Verhoeven

ELG

, et al. Vascular access animal models used in research. Ann Anat 2019; 225: 65–75.

18.

Liang

Wang

Liang

, et al. Migration of smooth muscle cells from the arterial anastomosis of arteriovenous fistulas requires Notch activation to form neointima. Kidney Int 2015; 88: 490–502.

19.

Skartsis

Manning

Wei

, et al. Origin of neointimal cells in arteriovenous fistulae: bone marrow, artery, or the vein itself? Semin Dial 2011; 24: 242–248.

20.

Chen

Wong

, et al. Vein graft adaptation and fistula maturation in the arterial environment. J Surg Res 2014; 188: 162–173.

21.

Hammes

Importance of the endothelium in arteriovenous fistula outcomes. Am J Nephrol 2016; 44: 426–427.

22.

Kirkton

Santiago-Maysonet

Lawson

, et al. Bioengineered human acellular vessels recellularize and evolve into living blood vessels after human implantation. Sci Transl Med 2019; 11: pii: eaau6934.

23.

Chlupac

Filova

Bacakova

[Vascular prostheses: 50 years of advancement from synthetic towards tissue engineering and cell therapy]. Rozhl Chir 2010; 89: 85–94.

24.

Walpoth

Moller

[Tissue engineering of vascular prostheses]. Chirurg 2011; 82: 303–310.

25.

Owens

Gasper

Rahman

, et al. Vein graft failure. J Vasc Surg 2015; 61: 203–216.

26.

Wang

Chai

Lin

, et al. Mouse models of neointimal hyperplasia: techniques and applications. Med Sci Monit 2006; 12: RA177–RA185.

27.

Lee

Roy-Chaudhury

Advances and new frontiers in the pathophysiology of venous neointimal hyperplasia and dialysis access stenosis. Adv Chronic Kidney Dis 2009; 16: 329–338.

28.

Terry

Shiu

, et al. Neointimal hyperplasia associated with synthetic hemodialysis grafts. Kidney Int 2008; 74: 1247–1261.

29.

Zhao

Jourd’heuil Fl Xue

, et al. Dual function for mature vascular smooth muscle cells during arteriovenous fistula remodeling. J Am Heart Assoc 2017; 6: e004891.

30.

Kang

Grande

Hillestad

, et al. A new model of an arteriovenous fistula in chronic kidney disease in the mouse: beneficial effects of upregulated heme oxygenase-1. Am J Physiol Renal Physiol 2016; 310: F466–F476.

31.

Juncos

Grande

Kang

, et al. MCP-1 contributes to arteriovenous fistula failure. J Am Soc Nephrol 2011; 22: 43–48.

32.

Liang

Wang

Han

, et al. Chronic kidney disease accelerates endothelial barrier dysfunction in a mouse model of an arteriovenous fistula. Am J Physiol Renal Physiol 2013; 304: F1413–F1420.

33.

Boos

Loew

Weigand

, et al. Engineering axially vascularized bone in the sheep arteriovenous-loop model. J Tissue Eng Regen Med 2013; 7: 654–664.

34.

Schuman

Babu

Sirolimus-loaded polyurethane graft for hemodialysis access in sheep. Vascular 2008; 16: 269–274.

35.

Rivero

KTV

Escobar

Forero

SDG

, et al. New regenerative vascular grafts for hemodialysis access: evaluation of a preclinical animal model. J Invest Surg 2018; 31: 192–200.

36.

Benedetto

Lipkowitz

Madden

, et al. Use of cryopreserved cadaveric vein allograft for hemodialysis access precludes kidney transplantation because of allosensitization. J Vasc Surg 2001; 34: 139–142.

37.

Madden

Lipkowitz

Benedetto

, et al. Decellularized cadaver vein allografts used for hemodialysis access do not cause allosensitization or preclude kidney transplantation. Am J Kidney Dis 2002; 40: 1240–1243.

38.

Rotmans

JI.

Animal models for studying pathophysiology of hemodialysis access. Open Urol Nephrol J 2014; 7: 14–21.