Biomaterials in the Development and Future of Vascular Grafts

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

INVITED REVIEW

Biomaterials in the development and future of vascular grafts

Lian Xue, MD, PhD,a and Howard P. Greisler, MD,a,b,c Maywood, Ill

Recent developments in the field of tissue engineering have re-invigorated the quest for more suitable biomaterials that are applicable to novel cardiovascular devices, including small-diameter vascular grafts. This review covers both commercially available and relevant newly developed experimental materials, including elastic polymers (polyurethane), the biodegradable and bioresorbable materials, and the naturally occurring materials, focusing on their potential applications in the development of future vascular substitutes. (J Vasc Surg 2003;37:472-80.)

The search for vascular substitute materials has thus far cular conduit. The emergence of tissue-engineering been a half-century endeavor.1 The initial failure of materials technology has made the development of a novel biologi- such as metal, glass, ivory, silk, and nylon brought 2 important cally viable vascular substitute feasible, and it may prove to criteria into focus: thrombogenicity and durability. Research be the ultimate solution for small-diameter vascular graft- was thus directed at inert materials that minimally interact ing. The purpose of this review is to highlight currently with blood and tissue. Polyethylene terephthalate (PET, Da- used and experimental biomaterials and their potential cron) and expanded polytetrafluoroethylene (ePTFE) are the applications in the development of future vascular grafts, products of this research and are currently the standard bio- focusing on those used for conventional open vascular materials of prosthetic vascular grafts. Examined by means of reconstructions. decades of use, both Dacron and ePTFE grafts have been shown to perform well at diameters Ͼ6 mm, but neither CURRENT MATERIALS material has been suitable for small-diameter (Ͻ4 mm) As aforementioned, the 2 standard polymers used for applications. Thus, finding a solution for small-diameter vascular grafts in clinical practice are Dacron and ePTFE. bypass grafting has become a major focus of attention. The Both PET and PTFE molecules are highly crystalline and mid- to long-term failure of existing synthetic grafts is hydrophobic, the 2 properties that prevent the polymers essentially caused by unfavorable healing processes, namely from hydrolysis. The hydrophobicity of the polymer has incomplete endothelialization and myointimal hyperplasia important implications in predicting surface interactions (IH). Seeking completely non-reactive substances is likely with blood and tissue. unrealistic. Optimizing tissue-biomaterial interactions to Dacron. PET was first introduced in 1939. DuPont elicit desirable results is thus a major emphasis of research. further developed it and patented its widely known Dacron Various modifications have been applied to Dacron and fiber in 1950.2 Vascular grafts made from Dacron were first ePTFE grafts to improve their function. Elastic polymers implanted by Julian in 1957 and DeBakey in 1958.1 have been used in the manufacture of compliant grafts on Clinically available Dacron grafts are fabricated in ei- the basis of the notion that compliance mismatch between ther woven or knitted forms. The multifilament Dacron the synthetic graft and native artery may contribute to IH. threads in woven grafts are fabricated in an over-and-under Biodegradable polymers can constitute a temporary scaf- pattern, which results in very limited porosity and minimal fold through which tissue ingrowth in vivo eventually re- creep of the finished graft. Knitted grafts are made with a places the prostheses and leave a complete biological vas- textile technique in which the Dacron threads are looped to create greater porosity and radial distensibility. The velour From the Department of Surgerya and the Department of Cell Biology, technique that extends the loops of yarn on the surfaces of b Neurobiology, and Anatomy, Loyola University Medical Center, and the fabrics has been used in an attempt to increase tissue Surgical Service, Hines VA Hospital.c Competition of interest: nil. incorporation. A crimping technique is used to increase the Reprint requests: Howard P. Greisler, MD, Loyola University Medical flexibility, distensibility, and kink-resistence of textile Center, Department of Surgery, 2160 South First Ave, Maywood, IL grafts. Prosthetic rings or coils are applied to the external 60153 (e-mail: [email protected]). surface of the grafts as external support to resist kinking and Copyright © 2003 by The Society for Vascular Surgery and The American possible mechanical compression. Association for Vascular Surgery. 0741-5214/2003/$30.00 ϩ 0 The high porosity of the knitted graft necessitates pre- doi:10.1067/mva.2003.88 clotting as a means of preventing transmural blood extrav- 472 JOURNAL OF VASCULAR SURGERY Volume 37, Number 2 Xue and Greisler 473 asation. Gelatin (Vascutek, Renfrewshire, Scotland), colla- untreated ePTFE graft of 56%, 46%, and 42%, respective- gen (Boston Scientific, Oakland, NJ), and albumin (Bard ly.19 The significance of heparin bound to synthetic grafts Cardiovascular, Billerica, Mass) are used to seal knitted will be further discussed in this review. Dacron graft pores. The gelatin and collagen in the Vas- Expanded polytetrafluoroethylene. PTFE was pat- cutek and Boston Scientific grafts are cross-linked by low ented by DuPont in 1937 as Teflon. Because of its partic- concentrations of formaldehyde, a method that results in a ular relatively inert characteristics, it was considered to be weak linkage that allows the gelatin or collagen to be an ideal electrical insulator.2 Its medical use began with its degraded in the body in Ͻ2 weeks.3,4 Bard uses glutaral- application in artificial heart valves in the early 1960s. In dehyde to cross-link albumin, and the albumin is absorbed 1969, Gore patented expanded ePTFE (Gore-tex), which in 2 months.5 is the material used in vascular grafts. The expanded poly- Dacron has a good stability and can persist for more mer is manufactured by means of a heating, stretching, and than 10 years after implantation without significant deteri- extruding process that produces a microporous material oration. However, knitted Dacron grafts have been prone more supportive of firm tissue adhesion. to dilate when implanted into the arterial environment, The PTFE molecule is biostable, and the graft made more because of fabrication technique than the polymer from it does not undergo biological deterioration within itself.6,7 Direct etiological association between graft dila- the body. The surface of the graft is electronegative, which tion and the later clinical complications has been rare.8 minimizes its reaction with blood components. ePTFE Other than this, there are no clinical differences grafts in grafts are manufactured by means of stretching a melt- complications and graft patency between woven and knit- extruded solid polymer tube, which then cracks into a ted grafts in their use as aortoiliac bypass grafts.9 Five-year non-woven porous tube. The characteristic structure of patency rates are 93% for aortic bifurcation grafts,10 but ePTFE is a node-fibril structure in which solid nodes con- only 43% for above-knee femoropopliteal bypass grafts,11 nect through fine fibrils, with an average internodal dis- and even lower for below-knee grafts. tance of 30 ␮m for a standard graft. Blood and tissue reactions to implanted grafts start Like Dacron grafts, ePTFE grafts perform well as aortic immediately after the restoration of circulation. The first substitutes, with a 5-year primary patency rate of 91% to step is a dynamic protein adsorption/desorption to syn- 95%.10,17 When used for femoropopliteal bypass grafting, thetic material surfaces, known as the Vroman effect,12 the 3- and 5-year patency rates are only 61%20 and 45%,11 followed by platelet adhesion, inflammatory cell infiltra- respectively, whereas the autogenous vein grafts have 5- tion, and endothelial cell (EC) and smooth muscle cell and 10-year cumulative patency rates of 77%21 and 50%, (SMC) migration.13,14 A coagulum containing fibrin, respectively.22 platelets, and blood cells builds up during the first few The initial host response to ePTFE grafts is similar to hours to days and stabilizes in a period of 6 to 18 months, that of Dacron grafts.13-15 A fibrin coagulum or amor- forming a compacted layer.15 The histological characteris- phous platelet-rich material develops in a time sequence tics observed within Dacron grafts is a compact fibrin layer that is similar in both materials. Lack of luminal surface on the blood-contacting surface and densely packed for- cellular coverage can be found at the midgraft region years eign body giant cells between the outer layer of the graft after human implants.23-25 In the outer wrap-reinforced wall and surrounding connective tissue capsule. The fibrin graft, the wrap limits the infiltration of the cells from layer within the midgraft remains acellular, regardless of perigraft tissue and leaves acellular fibrin matrix inside the whether the grafts are woven or knitted. An external velour graft wall.15 The densely fabricated wrap is manufactured surface permits more extensive and firmer incorporation of on the outer surface of some of the Gore-tex grafts as a the graft into surrounding tissue, but the function of an reinforcement to the graft wall. This wrap was beneficial in internal velour structure remains controversial, with the reducing post-implantation dilation. However, with the suggestion that it may enhance firm anchorage of the newer manufacturing technologies currently used, the wrap fibrin/platelet pseudointima.14,15 Protein impregnation is felt by many investigators to be unnecessary. changes the surface properties of Dacron grafts and may Several modifications to the basic graft have been pro- induce more inflammatory reaction, but it does not change posed for improving its function. One is to increase the the clinical patency rates of these grafts.3,16,17 graft permeability on the basis of the notion that the rate of A heparin-bonded Dacron graft by InterVascular (La tissue ingrowth is associated with graft porosity (in limited Ciotat, France) is currently available on the European porosity ranges) and that transmural capillary ingrowth can market. The heparin is bound primarily through Van der provide the cell source for the surface endothelialization. In Waals bonds to the polyester fiber that is pretreated with a a baboon model, enhanced tissue ingrowth with complete cationic agent, tridodecil-methyl-ammonium chloride endothelialization occurred in ePTFE grafts with a 60-␮m (TMAC). The external third of the graft wall is coated with or 90-␮m internodal distance, but the 90-␮m internodal collagen to prevent blood extravasation.18 In a comparative graft demonstrated focal areas of neointimal desquamation clinical trial involving 209 patients undergoing femoropop- at late periods.26 Increased tissue ingrowth and EC cover- liteal bypass grafting, the heparin-bonded Dacron graft age and higher patency rates of the high-porosity ePTFE exhibited slightly better patency rates at 1, 2, and 3 years of have also been reported in canine models.27,28 However, a 70%, 63%, and 55%, respectively, compared with rates of a human trial with high-porosity ePTFE failed to show any JOURNAL OF VASCULAR SURGERY 474 Xue and Greisler February 2003 advantage in platelet deposition as compared with standard grafts, the concept may also apply to scaffolds, which will be 30-␮m internodal distance ePTFE grafts.29 discussed. Another modification has focused on the luminal sur- More experimental modifications have been reviewed face of the graft. Carbon coating is used to increase the elsewhere.14 surface electronegativity so as to diminish thrombus forma- Despite the difference in both chemical and physical tion. Early studies demonstrated decreased platelet deposi- properties between ePTFE and Dacron grafts, the patency tion on carbon-coated grafts, but the overall patency rates rates are comparable at all positions.10,11,37,38 Little or only were not improved when compared with those of uncoated marginal clinical improvement has been achieved from grafts.28,30 A prospective multicenter clinical study, con- various modifications of the basic grafts. sisting of 81 carbon-impregnated ePTFE and 79 standard Elastic polymer—polyurethane. Both ePTFE and ePTFE grafts for below-knee popliteal and distal bypass Dacron grafts are relatively non-compliant. The compli- grafting, showed no difference in patency rates between the ance mismatch has been thought to contribute to the 2 groups as long as 2 years after implantation.31 However, development of IH at the anastomotic regions.39 Elastic a recent report on a multicenter trial in Europe involving polymers have been introduced to create radially compliant 128 carbon-coated ePTFE and 126 standard ePTFE grafts vascular grafts. for infrainguinal bypass grafting demonstrated the signifi- Polyurethanes (PUs) were originally developed com- cantly greater 1- and 2-year patency rates of the carbon- mercially in Germany in the 1930s as surface coatings, coated versus the standard grafts by means of life-table foams, and adhesives.40 Segmented PUs are copolymers analysis.32 comprising 3 different monomers, a hard domain derived The attachment of anticoagulant or antithrombotic from a diisocyanate, a chain extender, and a soft domain, agents to the graft has also been explored. The most most commonly polyol. The soft domain is mainly respon- investigated is heparin binding. Heparin-bound ePTFE sible for flexibility, whereas the hard domain imparts grafts demonstrated reduced thrombogenicity and im- strength. The selection of the 3 monomers can produce proved patency rates at 8 weeks compared with the stan- materials with different mechanical characteristics, which dard graft in the rat infrarenal aortic position.33 Whether makes PU an attractive biomaterial. Lycra is the trade name the anticoagulation works through continuous release of of a segmented polyether PU that was commercialized in heparin from the material that establishes an effective con- 1962 by DuPont. centration at the interface between blood and the graft As a biomaterial, PU was first used in manufacturing surface or through non-consumptive mechanisms of active implantable roller pumps and left ventricular assist devices function of the heparin immobilized on the material surface and as a coating for early artificial hearts.41 The superior is unclear. A major concern with the administration of elastic and compliant mechanical properties and acceptable heparin on the graft surface is the duration of heparin biocompatibility of PU make it an appealing material for function. Premature release or disturbance of functional vascular grafts. Developing PU-based small-diameter vas- heparin or the presence of a physical barrier because of cular grafts has attracted great interest from industry. adherent blood components implies a theoretical inefficacy The first generation of PU vascular grafts was devel- of the approach. In a recent report by Fisher et al from Gore oped with polyester PUs, which resulted in devices such as & Associates,34 heparin was covalently linked to a pre- Vascugraft by B. Braun Melsungen AG (Melsungen, Ger- treated bioactive surface of ePTFE grafts, which were then many). Although the initial report demonstrated good implanted into canine aortoiliac arteries. The surface hep- biocompatibility,42 the graft underwent surface chemical arin activity measured by means of antithrombin III uptake modification and deterioration in vivo.43,44 A clinical trial per unit area was 24.7 Ϯ 7.9 pmol/cm2 at 2 weeks and with Vascugraft for below-knee bypass grafting was aborted remained at 15.3 Ϯ 3.7 pmol/cm2 at 12 weeks. Although after 8 of 15 grafts had occluded in the first year.44 It has it seems promising, the actual benefit of this treatment been reported that PUs with polyester polyols as soft seg- needs to be proved in longer-term in vivo studies. ments are hydrolytically unstable.45 Various bioactive substances have been integrated onto Polyether-based PUs, such as in the Pulse-Tec (Newtec synthetic grafts by means of a number of delivery methods Vascular Products of North Wales, UK) vascular access to modulate the graft healing process. One example is fibrin graft, were then used. Polyetherurethane was relatively glue (FG) delivery of growth factors onto the ePTFE grafts. insensitive to hydrolysis but susceptible to oxidative degra- The growth factors can be slowly released from the FG, dation.45 The Pulse-Tec graft suffered from in vivo biodeg- retaining their bioactivities in vivo. ePTFE grafts impreg- radation and died in the product pipeline. Vectra (Thoratec nated with FG containing fibroblast growth factor (FGF)-1 Laboratories Corporation, Pleasanton, Calif) is another and heparin that were implanted into canine bypass graft vascular access graft made with polyetherurethaneurea. The models elicited greater endothelialization and tissue incor- graft is manufactured with an average pore size of 15 ␮m poration than untreated or FG/heparin (no FGF) treated and a non-porous layer under the luminal surface, which grafts.35,36 Many growth factors, such as FGF-2, platelet- makes it impervious to liquids.46 In a multicenter trial derived growth factor, and vascular endothelial growth involving 142 patients receiving either Vectra or ePTFE factor, have been tested by using various delivery systems.14 vascular access grafts with a follow-up time as long as 12 Although most of the earlier studies were done on synthetic months, no difference was found in the patency or compli- JOURNAL OF VASCULAR SURGERY Volume 37, Number 2 Xue and Greisler 475 cation rates of the 2 grafts, but the Vectra grafts allowed foam might degrade and form 2,4-toluene diamine, which earlier access.47 However, it was noted that the PU graft has been shown to cause liver cancer in laboratory animals.2 elongated with time after implantation, and the incidence The extent to which the initial compliance may affect of pseudointimal formation near the anastomosis was the long-term function of the graft remains controversial.59 higher than that in the ePTFE grafts. The Vectra graft It has long been realized that fibrous tissue formation received Food and Drug Administration (FDA) clearance within and surrounding an implanted graft would compro- in 2000. A small-diameter coronary bypass graft created by mise graft compliance. In a follow-up study of 8 patients the company with the same material is currently undergo- with iliofemoral artery woven Dacron grafts, the average ing clinical trial. graft diameter variation during the cardiac cycle was 6% at 1 A new generation of PU grafts uses polycarbonate- month after implantation, and it decreased to 1% after 1 based PUs that eliminate most ether linkages and thus are year.60 The mechanical behavior of vascular grafts in vivo is hydrolytically and oxidatively stable and more resistant to governed not only by the properties of the implanted graft, biodegradation.48 A non-woven poly(carbonate)urethane but also by the nature and the amount of tissue incorpora- graft, fabricated with a spray phase-inversion technique, tion. showed no significant degradation for as long as 6 months 49 in rat aorta. The graft demonstrated faster endotheliali- PRE-CLINICAL INVESTIGATIONAL zation, early stabilization of neointimal proliferation, and a BIOMATERIALS FOR VASCULAR GRAFTS AND thinner neointima compared with ePTFE grafts. The Cor- TISSUE-ENGINEERING SCAFFOLDS vita graft (Corvita, Miami, Fla), comprising a porous poly- carbonate PU inner tube filled with a glutaraldehyde cross- Biodegradable and bioresorbable polymers linked gelatin-heparin matrix reinforced on the outside Tissue-biomaterial interactions, which ultimately often with knitted Dacron mesh, displayed no signs of aneurysm result in graft failure, are inevitable as long as the prosthesis at 1 year after implantation in canine femoral arteries.50 remains implanted. Bioresorbable polymers possess the ad- When compared with ePTFE grafts, the PU graft overall vantage of leaving behind no prosthetic materials to keep showed no appreciable difference in neointimal formation stimulating persistent foreign-body reactions. Biodegrad- in the canine aortic model.51 The graft made of poly(car- able polymers, however, undergo fragmentation with ex- bonate-urea)urethane (Chronoflex, CardioTech Interna- posure to biological environments, which results in smaller tional, Woburn, Mass) is expected to have better stability degradation products that may or may not remain present because the polymer has no ether/ester linkages. The good either at the implantation site or in distant locations, such as stability of the graft was indicated both in vitro and in the lymphatic system.61 Theoretically, it is possible to tis- vivo.52-54 In a small animal study, the grafts were implanted sue-engineer a “neoartery,” assuming there is adequate in aortoiliac arteries in 4 dogs for 36 months.55 No evi- load bearing to resist dilation and include cellular compo- dence of polymer degradation or graft deterioration was nents with desirable physiologic characteristics. found. Histologically, there was IH in midgraft regions and The 2 most investigated bioresorbable polymers are around anastomoses, cellular infiltration and collagen dep- polyglycolic acid (PGA) and polylactic acid (PLA). PGA is osition inside the wall, and a fibrous capsule on the outer highly crystalline and is hydrophilic. It was used to make the surface of the graft with, reportedly, no foreign body reac- first synthetic absorbable suture. The suture loses its me- tion. The graft is currently undergoing clinical trial. Car- chanical strength 2 to 4 weeks after implantation because of boxylated PU treatment of the graft can create a surface in vivo hydrolytic degradation of the polymer.62 with reactive carboxylic acid groups to which hirudin has PLA is more hydrophobic than PGA because of the been covalently bound.56 The antithrombin activity of presence of an extra methyl group in the lactide molecule, immobilized hirudin may be expected to improve the graft which limits the water uptake and results in a lower hydro- performance. lysis rate. Lactic acid is a chiral molecule that, therefore, Tissue reactions to PU grafts are discrepant in the exists in 2 stereoisomeric forms, D-PLA and L-PLA. L- literature because factors such as different compositions of PLA is semicrystalline with high mechanical strength, and polymers, graft fabrication, porosity, and surface modifica- its hydrolytic product is naturally occurring L-lactic acid. tions all affect the results.56-58 No conclusion can be made Thus it is more frequently used in scaffold design. at this point as to whether PU grafts may be functionally The copolymers of glycolic acid and lactic acid mar- superior to ePTFE or Dacron grafts until more data be- keted under the trade name Vicryl and polyglactin 910 come available. (PG910) are widely used in the medical field as absorbable There have also been attempts at using PU with other sutures, orthopedic devices, and drug-delivery systems.62 biodegradable materials in the manufacture of biodegrad- Biodegradation, tissue regeneration, and mechani- able vascular grafts, which will be discussed. One major cal strength. A fully bioresorbable vascular graft made concern about PU grafts is the potential carcinogenic effect from Vicryl sheets was investigated in 1979.57 These early of its degradation products. In 1991, the FDA terminated grafts were prone to aneurysmal dilation and rupture. the use of PU foam as a surface-coating material for breast Grafts composed of woven PGA have been evaluated in a implants, after it had been marketed for Ͼ20 years. A rabbit model in our laboratory.63,64 An inner capsule com- statement issued by the FDA suggested that the implanted posed of a confluent layer of ECs and smooth muscle-like JOURNAL OF VASCULAR SURGERY 476 Xue and Greisler February 2003 myofibroblasts amid dense collagen fibers was formed 4 the compliance increased with the degradation of the weeks after implantation. Macrophage infiltration and PEG/PLA components in vitro. Yet, when implanted in phagocytosis were in parallel with the resorption of the vivo, the compliance reduced 20% after 12 weeks.70 PGA, which no longer could be identified within 3 months. Scaffolds for in vitro tissue engineering. After initial In this initial experiment, 10% of the PGA grafts showed attempts at directly implanting bioresorbable grafts that are aneurysmal dilation, with no difference between 1 to 3 totally dependent on tissue ingrowth in vivo, cell seeding months and 3 to 12 months, which suggests that the critical onto bioresorbable scaffolds has been exploited to initiate time for the development of aneurysms is during prosthetic functional tissue regeneration. SMC seeding onto PU/ resorption, before the ingrowth of tissue with adequate PLA scaffolds was found to enhance neomedia generation strength to resist hemodynamic pressures. and optimal media cell orientation.58 A Harvard-Massa- Polydioxanone (PDS), a material used clinically in bone chusetts Institute of Technology group led by Langer and pins and suture clips, is a more slowly resorbed compound. Vacanti constructed a tubular scaffold with woven polygla- Grafts made of PDS showed similar endothelialization of ctin as an outer layer and non-woven PGA as an inner the regenerated luminal surface after implantation, and layer.71 Autologous cells with mixed population from arte- PDS remained present for as long as 6 months. The ex- rial explants were seeded onto the scaffolds. After 7 days of planted specimens of these PDS grafts were able to with- in vitro culture, the constructed vessels were implanted into stand mean static bursting pressures of 6000 mm Hg and ovine pulmonary arteries. All 7 were patent for as long as 12 2000 mm Hg without fatigue.65 weeks. The polymer scaffold was replaced by cells and A critical feature of bioresorbable grafts is that they extracellular matrix (ECM) with time. However, the vessels must regenerate a tissue complex of sufficient strength demonstrated an increase in diameter. They then designed before loss of prosthetic integrity to minimize the possibil- a more durable PGA/polyhydroxyalkanoate (PHA) scaf- ity of aneurysmal dilation as a requirement for clinical fold.72 The inner layer was made of non-woven PGA efficacy. designed to degrade in 6 to 8 weeks, and the outer layer was Grafts constructed from Ն2 bioresorbable polymers made of nonporous PHA. PHAs are naturally occurring with different resorption rates or from bioresorbable poly- polyesters produced by several microorganisms.62 They can mers with a non-resorbable material component have been be degraded by hydrolysis, but have a degradation time of designed to dictate mechanical strength considerations. years. The PHA homopolymers are highly crystalline, rela- The woven grafts composed of yarns of 74% PG910 and tively hydrophobic, and usually extremely brittle. Copoly- 26% PDS demonstrated a 100% 1-year patency rate, with mers such as hydroxybutyrate with hydroxyvaleric acid are no aneurysms in the rabbit aorta model.66 The PG910 was less crystalline and more flexible. The PHA used in this totally resorbed within 2 months, and the PDS was totally study was a copolymer of polyhydroxyoctanoate (PHO) resorbed within 6 months. The regenerated arteries with- and hydroxyhexanoic acid and has a high tensile set of 35% stood 800 mm Hg of pulsatile systolic pressure ex vivo after 100% elongation. The constructed scaffolds had good without bursting. Partially resorbable grafts containing tensile strength, flexibility, and handling. Using this scaf- 69% PG910 and 31% polypropylene or 70% PDS and 30% fold, the authors showed that all the tissue-engineered polypropylene were implanted into canine aortoiliac arter- vessels were patent, with no aneurysms for as long as 150 ies. The overall patency rate was 90% for PG910/polypro- days after implantation into ovine abdominal aorta. The pylene and 86% for PDS/polypropylene for as long as 1 PGA layers were completely replaced by tissue within 3 to 4 year, with no aneurysms.67 PG910 was totally resorbed months. Development of endothelium and of a media, within 2 months, and PDS was totally resorbed within 4 containing collagen with the presence of elastin fibers, was months. Both grafts elicited tissue ingrowth, which re- evident.72 mained histologically stable from 4 months through 1 year. Organized tissue can only be generated in appropriate Polypropylene was chosen as the non-resorbable compo- mechanical conditions. Culturing SMC-seeded PGA scaf- nent because Dacron was found to inhibit the arterial folds with pulsatile flow for 8 weeks results in organization regeneration stimulated by the resorbable component.67 of SMCs into multilayer structures with orientated collagen Grafts prepared from a mixture of 5% PLA and 95% PU fibrils between cells. The vessel structure displayed a con- were evaluated in rat aorta.68,69 The grafts formed neointi- tractile response to vasoconstrictors, although the magni- mas in 6 weeks and neomedias with elastic laminae in 12 tude was only 15% to 20% of that of the native artery.73 The weeks after implantation, but aneurysmal dilatation devel- ECM accumulates after exposure to in vivo hemodynamic oped in 3 of 8 grafts, and another 2 dilated after 1 year. environments. The content of elastin and proteoglycans PU/PLA lattices started to disintegrate on day 12 and was demonstrated by means of biochemical analysis to peak completely fragmented within 1 year. The authors sug- at 8 and 16 weeks after implantation, respectively, after gested that a relatively compliant scaffold was necessary to exceeding their native artery levels and then decreased, induce circumferential orientation of SMCs. As aforemen- approaching that of native artery. Nevertheless, collagen tioned, the initial graft compliance can be compromised by content continuously increased to approximately 5 times tissue ingrowth in vivo. A graft comprised of a poly- that of the native artery within 24 weeks, without decline.74 etherurethane scaffold and sealed with polyethylene glycol ECM deposition is necessary for the establishment of graft (PEG)/PLA copolymer exhibited good compliance, and strength, but excessive matrix formation indicates unfavor- JOURNAL OF VASCULAR SURGERY Volume 37, Number 2 Xue and Greisler 477 able tissue remodeling. Much still needs to be learned to Type I collagen. Type I collagen is a major compo- control this balance. nent of most connective tissues and is present throughout The PGA polymer does not possess cell-anchoring the arterial wall. Its native state is resistant to most proteases sites. Surface modifications have been investigated to facil- but is readily degraded by a wide variety of proteases once itate cell attachment, spatial cell distribution, or both. denatured.80 Increasing intermolecular cross-links among Treatment with 1N NaOH transforms ester groups on the its 3 consisting peptide subunits can increase the tensile surface of PGA fibers to carboxylic acid and hydroxyl strength of collagen fibers and make it less susceptible to groups. The resultant hydrolyzed surface increased its ad- degradation. It has long been recognized that collagen, sorption of serum proteins and doubled seeded SMC at- with its integrin-binding domains, facilitates cell attach- tachment density.75 Incorporation of the RGD sequence to ment and that a collagen matrix can support tissue growth. the polymer surface can direct receptor-mediated cell ad- Because of its unique biological and physical properties, hesion.76 Patel et al77 synthesized a biotinylated PLA-PEG collagen has been extensively used in tissue-engineering (polyethylene glycol) copolymer. Biotinylated-RGD pep- applications. tide was immobilized on the polymer surface by avidin. The first complete tissue-engineered biological blood With patterning technology, the authors were able to conduit was constructed in 1986, with collagen gel as the achieve a controlled directed cell distribution. ECs adhered scaffold.81 Cultured bovine SMC and fibroblasts were sep- and spread only on the RGD-functionalized lines, sepa- arately embedded in collagen gel and assembled to form rated by no cell zones in between. This technique and the media and adventitia. ECs were seeded to the luminal concept of controlling specific cell distribution represent surface to form a monolayer of endothelium. Although the new possibilities in the tissue engineering field. graft obtained 92% EC coverage on the inner surface and It has been noticed that SMCs in proximity to residual longitudinal SMC organization, it failed to show the req- PGA fragments display an undifferentiated phenotype that uisite mechanical strength, even when reinforced with Da- is evidenced by a high mitotic rate and low expression on cron meshes. L’Heureux et al in 1993 modified this model contractile proteins.58 Degradation of polymer can pro- by using human umbilical vein ECs and SMCs and human duce acidic products and create a low pH microenviron- skin fibroblasts.82 They encountered the same mechanical ment, which stimulates chronic inflammation and induces limitation. To improve the mechanical strength of the fibrocollagenous tissue formation that impairs the compli- collagen constructs, several approaches have been used. ance of the graft and eventually may cause graft failure. Appropriate culture media and mechanical conditioning of Synthetic protein-based polymer. Synthetic protein- the constructed conduit stimulate its histological organiza- based polymers, such as elastic protein-based polymer, tion and improve its mechanical strength.83-85 Fabricating represent a new class of biomaterial. They are produced by collagen/elastin fibers into scaffolds with techniques such means of recombinant DNA technology and are biocom- as electrospinning instead of by using a collagen hydrogel patible and biodegradable. A model polymer is poly represents a new approach that may eventually be able to (GVGPV), the core sequence of which is a highly conserved provide enhanced scaffold strength.86 repeating sequence in elastin. Poly (GVGVP) cross-linked Decellularized biological scaffolds. To obtain a with ␥-irradiation exhibited an elastic modulus that was physiologic matrix scaffold resembling that of the native similar to the femoral artery.78 Degradation of the polymer artery, decellularized native vessel was introduced.87-90 can be achieved by the incorporation of carboxyamides The cells with their surface antigens are removed by means containing amino acids such as asparagine and glutamine of detergent and enzymatic extraction methods, leaving a into the chain. The carboxyamides hydrolyze to form car- well-preserved acellular matrix that provides a scaffold for boxylates, resulting in polymer breakdown. Depending on autologous cell ingrowth and allows favorable tissue re- the preceding and following amino acids, the degradation modeling. Allogenic scaffolds have achieved minimal im- can occur at times ranging from days to years. Conse- munoreactivity and good durability for as long as 6 years, quently, chemical clocks can be introduced to control the without aneurysmal degradation.87 Initially, patency was polymer degradation rate. Cell attachment can be achieved reported in 15 of 16 implants at 3 days to 6 years in canine by means of incorporating RGD sequences into the poly- femoral and carotid arteries,87 but a later study showed 5 of mer. It has been shown that ECs among other cell types can 9 implants failed from acute occlusion in coronary bypass attach to polymer poly(40[GVGVP], [GRGDSP]), spread, grafts, and only limited cellular repopulation occurred in a and grow to confluence.79 Both remarkable elasticity and follow-up period of 6 months.88 Detergent was eliminated controllable degradation make elastin-based polymers po- from the process because of a concern that its remnant may tentially desirable materials for scaffolds in blood vessel be cytotoxic. The difficulties with the supply of human tissue engineering. materials impose a significant limitation on allogeneic scaf- folds. In this respect, xenogeneic materials have an advan- Naturally occurring materials tage. However, unlike allogeneic grafts in which the ECMs The advantage of synthetic materials is that their mi- bear little antigenicity, interspecies matrix immunogenicity crostructure, strength, and speed of degradation can be exists for xenografts.89 Xenogeneic acellular scaffolds elicit controlled during production; natural materials, however, significant chronic immunoresponsive inflammation, facilitate cell repopulation and tissue remodeling. which is sufficient to destroy elastin structures.90,91 It is JOURNAL OF VASCULAR SURGERY 478 Xue and Greisler February 2003 necessary to either remove or mask the antigens from standing of biological reactions to vascular grafts with the structural proteins for xenogeneic sources to be used in principles of tissue engineering and innovations of technol- vascular tissue engineering. ogy to develop a new generation of vascular substitutes. A Autologous EC seeding was expected to address both living vascular graft with predictable and desirable biolog- the thrombogenicity of exposed collagen and the remain- ical functions will likely be constructed by culturing blood ing antigenicity of the matrix, but the actual results have vessel cells on biological/synthetic scaffolds in bioreactors been disappointing.92,93 When decellularized allogeneic with optimal hemodynamic and biomechanical conditions porcine carotid matrix seeded with autologous ECs was and supplemented with spatially and temporally controlled tested in carotid arteries, 46% of the 2-cm-long grafts failed 3-dimensional delivery of bioactive agents, the use of ge- within 1 week. The remaining grafts had a patency rate of netic engineering techniques, or both. This may provide 71% at 4 months. The patent grafts had well-preserved the ultimate solution for the current dismal long-term collagen and elastin structures, EC coverage, myofibroblast patency rates of small-caliber synthetic grafts. ingrowth, and some degree of inflammatory reaction at explantation.93 Scaffolds constructed from decellularized porcine in- REFERENCES testine with cross-linked type I bovine collagen deposited 1. Hess F. History of (micro) vascular surgery and the development of on the luminal surface have also been tested. The constructs small-caliber blood vessel prostheses. Microsurgery 1985;6:59-69. reportedly provided the necessary mechanical and hemody- 2. Friedman DW, Orland PJ, Greco RS. Biomaterials: an historical per- namic properties at implantation, and the scaffold was spective. In: Ralph S. Greco, editor. Implantation biology: the host response and biomedical devices. Baco Raton (Fla): CRC press; 1994. p. cellularized and remodeled within 90 days in rabbit carotid 1-12. 94 artery bypass graft models. Implanted as canine femoral 3. Jonas RA, Ziemer G, Schoen FJ, Britton L, Castaneda AR. A new artery interposition grafts, the constructs demonstrated sealant for knitted Dacron prostheses: minimal cross-linked gelatin. J myofibroblast repopulation through the scaffold and in- Vasc Surg 1988;7:414-9. complete endothelial coverage. Eight of 9 grafts were 4. Scott SM, Gaddy LR, Sahmel R, Hoffman H. A collagen coated vascular prosthesis. J Cardiovasc Surg (Torino) 1987;28:498-504. patent for as long as 9 weeks, but with significant anasto- 5. Cziperle DJ, Joyce KA, Tattersall CW, et al. Albumin impregnated motic IH. The diameter reductions were 7% at midgraft vascular grafts: albumin resorption and tissue reactions. J Cardiovasc and 56% and 42% at proximal and distal anastomoses, Surg 1992;33:407-14. respectively. No aneurysmal dilation was documented, but 6. Nunn Db, Carter MM, Donohue MT, Hudgins PC. Postoperative non-infectious seromal cavities around the grafts developed dilation of knitted Dacron aortic bifurcation graft. J Vasc Surg 1990; 95 12:291-7. in all dogs at various points. No prediction can currently 7. Alimi Y, Juhan C, Morati N, Girard N, Cohen S. Dilation of woven and be made on the prospect of this approach. knitted aortic prosthetic grafts: CT scan evaluation. Ann Vasc Surg 1994;8:238-42. ENDOVASCULAR GRAFTS 8. Blumenberg RM, Gelfand ML, Barton EA, Bowers CA, Gittleman DA. Endovascular grafting to date has used variations of the Clinical significance of aortic graft dilation. J Vasc Surg 1991;14:175- 80. same class of biomaterials that are currently used for open 9. Quarmby JW, Burnand KG, Lockhart SJ, et al. Prospective randomized graft placement. However, a substantial effort is underway trial of woven versus collagen-impregnated knitted prosthetic Dacron to develop novel chemistries, biomechanics, and surface grafts in aortoiliac surgery. Br J Surg 1998;85:775-7. modifications for these devices, work that is beyond the 10. Friedman SG, Lazzaro RS, Spier LN, Moccio C, Tortolani AJ. A scope of this review. The biomaterials discussed in this prospective randomized comparison of Dacron and polytetrafluoroeth- review are mainly focused on conventional vascular grafts. ylene aortic bifurcation grafts. Surgery 1995;117:7-10. 11. Green RM, Abbott WM, Matsumoto T, et al. Prosthetic above-knee Most of the principles are applicable to endovascular grafts. femoropopliteal bypass grafting: five-year results of a randomized trial. However, certain distinctive aspects should be considered J Vasc Surg 2000;31:417-25. for endovascular graft development. For instance, endovas- 12. Vroman L, Adams AL. Identification of rapid changes at plasma-solid cular grafts require an extremely thin-wall design, which interfaces. J Biomed Mater Res 1969;3:43-67. enables them to be compressed to fit into a delivery sheath 13. Xue L, Greisler HP. Blood vessels. In: Lanza RP, Langer R, Vacanti J, editors. Principles of tissue engineering. 2nd edition. San Diego (Calif): or catheter. The porosity requirement for endovascular Academic Press; 2000. p. 427-46. grafts may not necessarily be the same as for conventional 14. Greisler HP. Characteristics and healing of vascular grafts. In: Callow grafts. The wall structure of the grafts may be affected by AD, Ernst CB, editors. Vascular surgery: theory and practice. Stamford deployment procedures. In addition, endovascular grafts (Conn): Appleton & Lange; 1995. p. 1181-212. are surrounded by blood clots and atherosclerotic lesions, 15. Davids L, Dower T, Zilla P. The lack of healing in conventional vascular grafts. In: Zilla P, Greisler HP, editors. Tissue engineering of vascular which may result in tissue reactions that are different from prosthetic grafts. Austin (Texas): R.G. Landes Company; 1999. p. that of the conventional grafts. 3-44. 16. De Mol Van Otterloo JC, Van Bockel JH, Ponfoort ED, et al. Systemic CONCLUSIONS effects of collagen-impregnated aortoiliac Dacron vascular prostheses ePTFE and Dacron are the standard materials for large- on platelet activation and fibrin formation. J Vasc Surg 1991;14:59-66. 17. Prager M, Polterauer P, Bo¨hmig HJ, et al. Collagen versus gelatin- diameter vascular grafts, but no ideal alternative to autolo- coated Dacron versus stretch polytetrafluoroethylene in abdominal gous vein grafts is currently available for small-diameter aortic bifurcation graft surgery: results of a seven-year prospective, applications. We are on the verge of integrating our under- randomized multicenter trial. Surgery 2001;130:408-14. JOURNAL OF VASCULAR SURGERY Volume 37, Number 2 Xue and Greisler 479

18. Lambert AW, Fox AD, Williams DJ, Horrocks M, Budd JS. Experience 41. Boretos JW, Pierce WS. Segmented polyurethane: a new elastomer for with heparin-bounded collagen-coated grafts for infrainguinal bypass. biomedical applications. Science 1967;158:1481-2. Cardiovasc Surg 1999;7:491-4. 42. Hess F, Jerusalem C, Steeghs S, et al. Development and long-term fate 19. Devine C, McCollum C. Heparin-bounded Dacron or polytetrafluoro- of a cellular lining in fibrous polyurethane vascular prostheses implanted ethylene for femoropopliteal bypass grafting: a multicenter trial. J Vasc in the dog carotid and femoral artery. J Cardiovasc Surg 1992;33:358- Surg 2001:33:533-9. 65. 20. Post S, Kraus T, Muller-Reinartz U, et al. Dacron vs polytetrafluoro- 43. Marois Y, Paris E, Zhang Z, et al. Vascugraft microporous polyester- ethylene grafts for femoropopliteal bypass: a prospective randomized urethane arterial prosthesis as a thoraco-abdominal bypass in dogs. multicentre trial. Eur J Vasc Endovasc Surg 2001;22:226-31. Biomaterials 1996;17:1289-300. 21. Taylor LM, Edwards JM, Porter JM. Present status of reversed vein 44. Zhang Z, Marois Y, Guidoin RG, et al. Vascugraft polyurethane arterial bypass grafting: five-year results of a modern series. J Vasc Surg 1990; prosthesis as femoro-popliteal and femoro-peroneal bypasses in hu- 10:220-5. mans: pathological, structural and chemical analyses of four excised 22. Donaldson MC, Mannick JA, Whittemore AD. Femoro-distal bypass grafts. Biomaterials 1997;18:113-24. with in situ greater saphenous vein. Ann Surg 1991;213:457-65. 45. Santerre JP, Labow RS, Duguay DG, Erfle D, Adams GA. Biodegrada- 23. Sottiurai VS, Yao JS, Flinn WR, Batson RC. Intimal hyperplasia and tion evaluation of polyether and polyester-urethanes with oxidative and neointima: an ultrastructural analysis of thrombosed grafts in humans. hydrolytic enzymes. J Biomed Mater Res 1994;28:1187-99. Surgery 1983;93:809-17. 46. Eberhart A, Zhang Z, Guidoin R, et al. A new generation of polyure- 24. Bellon JM, Bujan J, Contreras LA, Hernando A, Jurado F. Similarity in thane vascular prostheses: rara avis or ignis fatuus? J Biomed Mater Res behavior of polytetrafluoroethylene (ePTFE) prostheses implanted into 1999;48:546-58. different interfaces. J Biomed Mater Res 1996;31:1-9. 47. Glickman MH, Stokes GK, Ross JR, et al. Multicenter evaluation of a 25. Clowes AW, Gown AM, Hanson SR, Reidy MA. Mechanisms of arterial polytetrafluoroethylene vascular access graft as compared with the ex- graft failure. 1. Role of cellular proliferation in early healing of PTFE panded polytetrafluoroethylene vascular access graft in hemodialysis prostheses. Am J Pathol 1985;118:43-54. applications. J Vasc Surg 2001;34:465-72. 26. Clowes AW, Kirkman TR, Reidy MA. Mechanisms of arterial graft 48. Tanzi MC, Fare S, Petrini P. In vitro stability of polyether and polycar- healing: rapid transmural capillary ingrowth provides a source of intimal bonate urethanes. J Biomater Appl 2000;14:325-48. endothelium and smooth muscle in porous PTFE prostheses. Am J Path 49. Jeschke MG, Hermanutz V, Wolf SE, Ko¨veker GB. Polyurethane vascular prostheses decreases neointimal formation compared with ex- 1986;123:220-30. panded polytetrafluoroethylene. J Vasc Surg 1999;29:168-76. 27. Cameron BL, Tsuchida H, Connall TP, et al. High porosity PTFE 50. Wilson GJ, MacGregor DC, Klement P, et al. The composite Coreth- improves endothelialization of arterial grafts without increasing early ane/Dacron vascular prosthesis. Canine in vivo evaluation of 4 mm thrombogenicity. J Cardiovasc Surg 1993;34:281-5. diameter grafts with 1 year follow-up. ASAIO Trans 1991;37:M475- 28. Akers DL, Du YH, Kempczinski RF. The effect of carbon coating and M476. porosity on early patency of expanded polytetrafluoroethylene grafts: an 51. Akiyama N, Esato E, Fujioka K, Zempo N. A comparison of CORVITA experimental study. J Vasc Surg 1993;18:10-5. and expanded polytetrafluoroethylene vascular grafts implanted in the 29. Kohler TR, Stratton JR, Kirkman TR, et al. Conventional versus high- abdominal aortas of dogs. Surg Today 1997;27:840-5. porosity polytetrafluoroethylene grafts: clinical evaluation. Surgery 52. Salacinski HJ, Odlyha M, Hamilton G, Seifalian AM. Thermo-mechan- 1992;112:901-7. ical analysis of a compliant poly(carbonate-urea)urethane after exposure 30. Tsuchida H, Cameron BL, Marcus CS, Wilson SE. Modified polytetra- to hydrolytic, oxidative, peroxidative and biological solutions. Bioma- fluoroethylene: indium 111-labeled platelet deposition on carbon-lined terials 2002;23:2231-40. and high porosity polytetrafluoroethylene grafts. J Vasc Surg 1992;16: 53. Salacinski HJ, Tai NR, Carson RJ, et al. In vitro stability of a novel 643-9. compliant poly(carbonate-urea)urethane to oxidative and hydrolytic 31. Bacourt F. Prospective randomized study of carbon-impregnated poly- stress. J Biomed Mater Res 2002;59:207-18. tetrafluoroethylene grafts for below-knee popliteal and distal bypass, 54. Edwards A, Carson RJ, Szycher M, Bowald S. In vitro and in vivo results at 2 years. Ann Vasc Surg 1997;11:569-603. biodurability of a compliant microporous vascular graft. J Biomater 32. Groegler FM, Kapfer X, Meichelboeck W. Does carbon improve PTFE Appl 1998;13:23-45. bypass material? Proceedings of the 20th World Congress of the Inter- 55. Salacinski HJ, Tiwari A, Carson RJ, et al. In vivo biocompatibility and national Union of Angiology; 2002 April 7-11; New York. biostability of a novel compliant microporous poly(carbonate-urea)ure- 33. Walpoth BH, Rogulenko R, Tikhvinskaia E, et al. Improvement of thane vascular graft. Cardiovasc Pathol 2002;11:24. patency rate in heparin-coated small synthetic vascular grafts. Circula- 56. Phaneuf MD, Dempsey DJ, Bide MJ, et al. Bioengineering of a novel tion 1998;98:II319-II323. small diameter polyurethane vascular graft with covalently bound re- 34. Fisher JL, Thomson RC, Moore JW, Begovac PC. Functional parame- combinant hirudin. ASIAO J 1998;44:M653-M658. ters of thromboresistant heparinized e-PTFE vascular grafts. Cardiovasc 57. Bowald S, Busch C, Eriksson I. Arterial regeneration following polygla- Pathol 2002;11:42. ctin 910 suture mesh grafting. Surgery 1979;86:722-9. 35. Greisler HP, Cziperle DJ, Petsikas D, et al. Enhanced endothelializa- 58. Yue X, van der Lei B, Schakenraad JM, et al. Smooth muscle cell seeding tion of expanded PTFE grafts by heparin binding growth factor-type1 in biodegradable grafts in rats: a new method to enhance the process of pretreatment. Surgery 1992;112:244-55. arterial wall regeneration. Surgery 1988;103:206-12. 36. Gray JL, Kang SS, Zenni GC, et al. FGF-1 affixation stimulates ePTFE 59. Greenwald SE, Berry CL. Improving vascular grafts: the importance of endothelialization without intimal hyperplasia. J Res Surg 1994;57: mechanical and haemodynamic properties. J Pathol 2000;190:292-9. 596-612. 60. Gozna ER, Mason WF, Marble AE, Winter DA, Dolan FG. Necessity 37. Polterauer P, Prager M, Holzenbein T, et al. Dacron versus polytetra- for elastic properties in synthetic arterial grafts. Can J Surg 1974:17: fluoroethylene for Y-aortic bifurcation grafts: a six-year prospective, 176-9. randomized trial. Surgery 1992;111:626-33. 61. Shalaby S. Degradable materials: perspectives, issues and opportunities. 38. Johnson WC, Lee KK. Comparative evaluation of externally supported In: Barenberg S, Brash J, Narayan R, Redpath A, editors. The First Dacron and polytetrafluoroethylene prosthetic bypasses for femoro- International Scientific Consensus Workshop on Degradable Materials. femoral and axillofemoral arterial reconstructions. Veterans Affairs Co- Toronto: CRC Press; 1989. p. 678. operative Study #141. J Vasc Surg 1999;30:1077-83. 62. Pachence JM, Kohn J. Biodegradable polymers. In: Lanza RP, Langer 39. Salacinski HJ, Goldner S, Giudiceandrea A, et al. The mechanical R, Vacanti J, editors. Principles of tissue engineering. 2nd edition. San behavior of vascular grafts: a review. J Biomater Appl 2001;15:241-78. Diego (Calif): Academic Press; 2000. p. 263-77. 40. Batich C, DePalma D. Materials used in breast implants: silicones and 63. Greisler HP. Arterial regeneration over absorbable prostheses. Arch polyurethanes. J Long-term Effects Med Implants 1992;1:255. Surg 1982;117:1425-31. JOURNAL OF VASCULAR SURGERY 480 Xue and Greisler February 2003

64. Greisler HP, Kim DU, Price JB, Voorhees AB. Arterial regenerative book of vascular biology and diseases. 2nd edition. Boston: Little, activity after prosthetic implantation. Arch Surg 1985;120:315-23. Brown and Company; 1996. p. 2-67. 65. Greisler HP, Ellinger J, Schwarcz TH, et al. Arterial regeneration over 81. Weinberg C, Bell E. A blood vessel model constructed from collagen polydioxanone prostheses in the rabbit. Arch Surg 1987;122:715-21. and cultured vascular cells. Science 1986;231:397-400. 66. Greisler HP, Endean ED, Klosak JJ, et al. Polyglactin 910/polydiox- 82. L’Heureux N, Germain L, Labbe R, Auger FA. In vitro construction of anone bicomponent totally resorbable vascular prostheses. J Vasc Surg a human blood vessel from cultured vascular cells: a morphologic study. 1988;7:697-705. J Vasc Surg 1993;17:499-509. 67. Greisler PH, Tattersall CW, Klosak JJ, et al. Partially bioresorbable 83. L’Heureux N, Paquet S, Labbe R, Germain L, Auger FA. A completely vascular grafts in dogs. Surgery 1991;110:645-55. biological tissue-engineered human blood vessel. FASEB J 1998;12:47- 68. van der Lei B, Nieuwenhuis P, Molenaar I, Wildevuur CR. Long-term 56. biologic fate of neoarteries regenerated in microporous, compliant, 84. Ziegler T, Nerem RM. Tissue engineering a blood vessel, regulation of biodegradable, small-caliber vascular grafts in rats. Surgery 1987;101: vascular biology by mechanical stresses. J Cell Biochem 1994;56:204-9. 459-67. 85. Seliktar D, Black RA, Vitro RP, Nerem RM. Dynamic mechanical 69. van der Lei B, Wildevuur CR, Dijk F, et al. Sequential studies of arterial conditioning of collagen-gel blood vessel constructs induces remodel- wall regeneration in microporous, compliant, biodegradable small- ing in vitro. Ann Biomed Eng 2000;28:351-62. caliber vascular grafts in rats. J Thorac Cardiovasc Surg 1987;93:695- 86. Matthews JA, Stitzel JD, Wnek GE, Simpson DG, Bowlin GL. Smooth 707. muscle cell migration in electrospun poly(lactic acid) and collagen/ 70. Izhar U, Schwalb H, Borman JB, et al. Novel synthetic selectively elastin. Cardiovasc Pathol 2002;11:13. degradable vascular prostheses: a preliminary implantation study. J Surg 87. Wilson FJ, Yeger H, Klement P, Lee JM, Courtmant DW. Acellular Res 2001;95:152-60. matrix allograft small caliber vascular prostheses. ASAIO Trans 1990; 71. Shinoka T, Shum-Tim D, Ma PX, et al. Creation of viable pulmonary 36:M340-M343. artery autografts through tissue engineering. J Thorac Cardiovasc Surg 88. Wilson GJ, Courtman DW, Klement P, Lee JM, Yeger H. Acellular 1998;115:536-46. matrix: a biomaterials approach for coronary artery bypass and heart 72. Shum-Tim D, Stock U, Hrkach J, et al. Tissue engineering of autolo- valve replacement. Ann Thorac Surg 1995;60(2 Suppl):S353-S358. gous aorta using a new biodegradable polymer. Ann Thorac Surg 89. Allaire E, Bruneval P, Mandet C, Becquemin JP, Michel B. The 1999;69:2298-305. immunogenicity of the ECM in arterial xenografts. Surgery 1997;122: 73. Niklason LE, Abbott W, Gao J, et al. Morphologic and mechanical 73-81. characteristics of engineered bovine arteries. J Vasc Surg 2001;33:628- 90. Courtman DW, Errett BF, Wilson GJ. The role of crosslinking in 38. modification of the immue response elicited against xenogenic vascular 74. Stock UA, Wiederschain D, Kilroy SM, et al. Dynamics of ECM acellular matrices. J Biomed Mater Res 2001;55:576-86. production and turnover in tissue engineered cardiovascular structures. 91. Bader A, Steinhoff G, Strobl K, et al. Engineering of human vascular J Cell Biochem 2001;81:220-8. aortic tissue based on a xenogeneic starter matrix. Transplantation 75. Gao J, Niklason L, Langer R. Surface hydrolysis of poly(glycolic acid) 2000;70:7-14. meshes increases the seeding density of vascular smooth muscle cells. 92. Teeken OE, Bader A, Steinhoff G, Haverich A. Tissue engineering of J Biomed Mater Res 1998;42:417-24. vascular grafts: human cell seeding of decellularised porcine matrix. Eur 76. Drumheller PD, Hubbell JA. Polymer networks with grafted cell- J Vasc Endovasc Surg 2000;19:381-6. adhesion peptides for highly biospecific cell adhesive substrates. Anal 93. Teebken OE, Pichlmaier AM, Haverich A. Cell seeded decellularised Biochem 1994;222:380-8. allogenic matrix grafts and biodegradable polydioxanone-prostheses 77. Patel N, Padera R, Sanders GHW, et al. Spatially controlled cell engi- compared with arterial autografts in a porcine model. Eur J Vasc neering on biodegradable polymer surfaces. FASEB J 1998;12:1447- Endovasc Surg 2001;22:139-45 . 54. 94. Huynh T, Abraham G, Murray J, et al. Remodeling of an acellular 78. Urry DW, Pattanaik A. Elastic protein-based materials in tissue recon- collagen graft into a physiologically responsive neovessel. Nat Biotech- struction. Ann N Y Acad Sci 1997;831:32-46. nol 1999;17:1083-6. 79. Nicol A, Gowda DC, Urry DW. Cell adhesion and growth on synthetic 95. Nemcova S, Noel AA, Jost CJ, et al. Evaluation of a xenogeneic acellular elastomeric matrices containing arg-gly-asp-ser. J Biomed Mater Res collagen matrix as a small-diameter vascular graft in dogs—preliminary 1992;26:393-413. observations. J Invest Surg 2001;14:321-30. 80. Seyer JM, Kang AH. Connective tissues of the subendothelium. In: Loscalzo J, Creager MA, Dzan VJ, editors. Vascular medicine—a text Submitted Jul 30, 2002; accepted Aug 15, 2002.