| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Experimental Studies |
1 From the Depts of Radiology (H.J.C., D.F.K., W.F.M., J.K.M., J.E.D., M.E.J.), Neurological Surgery (S.B.H., G.A.H.), and Pathology (M.B.L.), University of Virginia Health Sciences Center, Charlottesville, and ReGen, Franklin Lakes, NJ (H.B.L., S.T.L.). From the 1997 RSNA scientific assembly. Received May 14, 1998; revision requested July 13; revision received May 11, 1999; accepted June 28. Supported in part by a 1995 (2) Seed Grant from the RSNA R&E Foundation. D.F.K. supported by the RSNA R&E Foundation as a 1997 RSNA scholar. Address reprint requests to H.J.C., Dept of Radiology, Emory University Hospital, 1364 Clifton Rd NE, Atlanta, GA 30322 (e-mail:harry_cloft@emory.org).
 |
Abstract |
|---|
 TOP Abstract IntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: A composite nitinol-collagen endovascular stent-graft with a 4-mm inner diameter was deployed in the abdominal aorta in nine rabbits. Angiography was performed, and the rabbits were sacrificed at 1, 2, and 7 days and at 1 and 3 months. The portion of the aorta containing the stent-graft was excised and was histologically evaluated.
RESULTS: All stent-grafts were patent at all time points. On days 1, 2, and 7 after implantation, scattered red and white blood cells adhered to the stent-graft. At 1 month, the stent-graft was endothelialized and was infiltrated with fibroblasts that deposited collagen within the interstices of the implanted collagen material. At 3 months, there was additional collagen deposition within the interstices of the stent-graft that did not narrow the lumen of the stent-grafts.
CONCLUSION: Type I collagen as a intravascular stent-graft material is biocompatible for at least 3 months in rabbits. It is rapidly endothelialized and does not cause reactive stenosis. As a versatile and biocompatible polymer, collagen is potentially useful in the construction of endovascular stent-grafts for use in human arteries.
Index terms: Animals • Aorta, grafts and prostheses, 981.1268 • Aortography, 981.121 • Interventional procedures, experimental studies, 981.1268 • Stents and prostheses, 981.1268
|
Introduction |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Although a great deal of effort and resources have been invested in the development of percutaneously placed stent-grafts in recent years, data regarding the endovascular biocompatibility of materials remain sparse. In the development of stent-grafts, the choice of stent-graft material, which is critical to the successful development of these devices, currently is largely influenced by regulatory issues, availability, and ease of fabrication (9).
Studies of stent-grafts that were placed into the large arteries of the aortoiliac system had promising results (1–3). However, the stent-grafts that were placed into small arteries (<5 mm) did not have high patency rates because of reactions to the stent-graft materials (4,10,11). Synthetic polymers, such as polyurethane (10) and polyethylene teraphthalate (including Dacron) (4,11), were shown to elicit severe inflammatory reactions that resulted in the occlusion of small arteries in the swine model. The accumulation of cellular and fibrous material on the surface of the implanted stent-grafts generally has little effect in large vessels, but they can cause substantial luminal narrowing and possible occlusion in vessels with small diameters. Thus, the development of biocompatible materials is critical to the success of using stent-grafts in small arteries.
A natural biologic polymer, such as collagen, is potentially more biocompatible than the synthetic polymers. Collagen also possesses sufficient tensile strength to exclude blood flow into an aneurysm at systemic blood pressures. A bovine type I collagen preparation was developed for use as a stent-graft material. Our purpose was to study the biocompatibility of this collagen as a material for small-vessel stent-grafts in a rabbit model.
|
MATERIALS AND METHODS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
To incorporate heparin into the bulk collagen preparation, the preparation was hydrated at pH 2.5 in the presence of heparin (Diosynth; Chicago, Ill; 197 U/mg). This technique resulted in as much as 5% (wt/wt) heparin in the final preparation. Collagen tubes with 4-mm inner diameters and 0.10–0.15-mm wall thicknesses were formed on a mandrel. The collagen fibers on the mandrel were freeze-dried and then were released from the mandrel for chemical cross-linking with formaldehyde vapor. The cross-linked collagen tubes were thoroughly degassed under a vacuum to remove the free formaldehyde. The tubes were cut to 1-cm lengths and were packaged for sterilization with 
radiation.
To keep the soft collagen tubes patent, they were attached to handmade, nitinol, single-body Z stents (Fig 1). The Z stents were constructed with 0.009-inch–diameter nitinol wire, were 8–10-mm long, and were self-expanding to a 4-mm diameter. The collagen tubes were attached, end to end, to the nitinol stents by using 6-0 polypropylene suture (Prolene; Ethicon, Somerville, NJ).
|
|
The stent-graft placement procedure was performed by using a sterile technique. Stent-grafts were deployed in nine female New Zealand White rabbits. Right femoral arterial cutdowns were performed, and the femoral artery was punctured with a 21-gauge Teflon sheath over a needle. A 0.018-inch–diameter guide wire was placed though the Teflon sheath, and the artery was sequentially dilated so that a 6-F sheath could be placed with the distal tip in the abdominal aorta.
The stent-graft was manually compressed and was pushed into an introducer. The collagen material was sufficiently strong to withstand this process without damage. The stent-graft was introduced into the sheath and was pushed to the distal end of the sheath by using a Teflon pusher (Fig 1). When the stent-graft was positioned correctly, it was kept in place with the pusher while the introducer sheath was withdrawn until the stent-graft was fully expanded.
The luminal diameters were not quantitatively measured but qualitatively assessed. The pusher was withdrawn, and nonionic contrast material (Omnipaque 240 [iohexol]; Nycomed Amersham, Princeton, NJ; 5 mL) was then injected through the sheath during fluoroscopic observation to confirm the position and complete expansion of the stent-graft. The sheath was then removed, the femoral artery was ligated, and the wound was sutured.
Follow-up Angiography and Histologic Examination
One animal was sacrificed at 1 day, one at 2 days, two at 7 days, two at 1 month, and three at 3 months after implantation. At the time of sacrifice, the animals were anesthetized, and left femoral arterial cutdowns were performed. The femoral artery was punctured with a 21-gauge Teflon sheath over a needle. The needle was removed, and the nonionic contrast material was injected through the Teflon sheath. The contrast material was forcefully injected to cause reflux into the abdominal aorta above the stent-graft while conventional radiography was being performed. The animal was then euthanized by using an overdose of pentobarbital sodium.
Pressure fixation (at 80–100 mm Hg) of the aorta was then immediately performed through the femoral sheath by using a 4% paraformaldehyde solution. The aortic segment containing the stent-graft was then carefully excised and placed into a 4% paraformaldehyde solution. The nitinol stent was carefully removed, and the segment of the artery containing the collagen tube was sent for histologic evaluation. Paraffin-embedded sections (10-µm) were obtained and were stained with hematoxylin-eosin for an analysis of the cellular content or with Masson trichrome stain for an evaluation of the collagen content. The histologic analysis was performed by using a light microscope.
|
RESULTS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
|
|
|
|
|
|
At 3 months after implantation, additional collagen was deposited within the interstices of the stent-graft, compared with the deposition in the specimens examined at 1 month (Fig 3e, 3f). Although there was collagen deposition, it did not narrow the lumen of the stent-graft. There was no evidence of an inflammatory rejection of the collagen material. In places where the stent-graft was not closely applied to the arterial wall, an organized blood clot was present between the collagen material and the arterial wall.
|
DISCUSSION |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Collagen is a highly thrombogenic material, as it is one of the primary initiators of the clotting cascade (12). However, the thrombogenicity of collagen is substantially reduced with aldehyde cross-linking (13). We further reduced the thrombogenic potential of our collagen material by binding heparin to it. Platelet aggregation occurs on collagen (14), yet heparin has been shown to reduce this on its surfaces (15). Platelet aggregation was evaluated in our collagen preparation. Heparin bound to its surfaces has also been shown to inactivate thrombin (16).
Although, to our knowledge, the use of collagen as an endovascular stent-graft material has not been reported previously, surgically anastomosed collagen vascular grafts have been described. Fibrocollagenous vascular grafts can be produced by implanting silicone rubber rods surrounded by a Dacron mesh into the subcutaneous tissues of animals and by harvesting the surrounding tubular fibrocollagenous scar weeks later (17–20). Such 0.8-mm–diameter grafts were anastomosed to the abdominal aorta in rats and remained patent for as long as 6 months (19), whereas 8-mm–diameter grafts that were placed into canine abdominal aorta were documented to have been patent for 2 years (17). A small number of these fibrocollagenous grafts were used in humans; documented patency was as long as 19 months. Patency rates for these fibrocollagenous grafts can be improved by binding urokinase to their inner surfaces (20).
Another type of collagen tube for use as a vascular graft can be made by removing the carotid arteries of dogs and other animals, by digesting them with chymotrypsin, and by tanning them with glutaraldehyde to obtain an arterial heterograft that consists mostly of collagen and elastic fibers (21,22). Binding heparin to these collagen grafts improves their patency rate (21,22). A tube made of extruded collagen that was glued with gelatin to a Dacron outer tube was bound with heparin and was placed into dogs and pigs for as long as 4
years and into humans for as long as 4 years (23).
The collagen preparation described herein is more pure and more versatile than that of the arterial grafts previously described (21,22) for use in endovascular prostheses. The collagen tube used in this study required a metal stent to maintain its patency since the collagen itself did not have enough radial force to hold itself open reliably. The end-to-end placement of the collagen tubes and the metal stent design were used because they allowed us to test the biocompatibility of the collagen, independent of that of the metal stent, as previously described by Marin et al (24).
For the treatment of vascular disease, collagen-covered stent-grafts could be manufactured by directly binding collagen to a metal stent and by covering all of the spaces between the struts. Stent-grafts could also be manufactured to have metal Z stents at each end to hold the stent-graft open, with collagen alone to bridge the space between the Z stents and to provide flexibility. Collagen-stent grafts could be rapidly developed for clinical use by simply modifying the existing metal stents and the designs for deployment systems.
The long-term patency of our collagen-covered stent-graft in small arteries stands in contrast to the disappointing results obtained with stent-grafts that are covered with various synthetic polymers. A Dacron-covered stent-graft (Cragg EndoPro; Mintec, Bahamas) has been commercially available since 1993. The Dacron-covered stent-graft was developed as a means of preventing neointimal proliferation (8). In a short-term experiment, Dacron-covered stent-grafts caused inflammatory reactions that led to occlusions in 5-mm swine carotid arteries (4). A possible explanation involves the local toxic reactions related to the residue from the materials used in the manufacturing process; these include heavy metals, nitrogen-containing aromatic compounds, antioxidants, and oligomers (4).
Polyurethane (10) and polyethylene teraphthalate (11) also cause severe inflammation and occlusion when they are used as stent-graft materials. Polytetrafluoroethylene, however, is a synthetic polymer that holds promise as a stent-graft material. At 1 year after implantation, 3-mm–diameter polytetrafluoroethylene stentgrafts had a 73% patency rate in atherosclerotic superficial femoral arteries in humans (25).
The purity of bovine collagen preparations is known to be important, as impurities have been implicated in the occasional inflammatory reactions to collagen-impregnated Dacron grafts (26). Collagen, a naturally occurring structural protein, is poorly antigenic. Therefore, it is not likely to induce an inflammatory response (13). Collagen-coated Dacron grafts have been used in humans for many years (27,28). The purpose of the collagen coating in these grafts is simply to seal the pores in the Dacron to make the graft impervious to blood at the time of implantation (27). While collagen has a limited role in the function of these grafts, the use of these grafts shows that bovine collagen can be successfully used in a human vascular prostheses.
Materials used in the production of stent-grafts may be permanent or biodegradable. Aldehyde cross-linking has been shown to greatly enhance the in vivo persistence of collagen implants (29). Adjustments in cross-linking can make collagen stent-grafts more or less biodegradable, depending on what is desired for a particular application. The data show that, for over 3 months, the collagen used in the stent-graft in the current study was not biodegradable. In addition, reducing the thickness of the collagen tube may also result in a more favorable biologic response. Improvements in the biologic response could include decreased deposition of new collagen.
Collagen stent-grafts could also be used as vehicles for the local delivery of drugs. Drugs such as heparin (21,22), urokinase (20), and plasmin (30) could be bound to the stent-grafts. In addition, naturally occurring growth factors that may have useful applications as local drugs can be bound to the stent-grafts. They can improve the endovascular stent-grafts by modifying the biologic response to it, or they can be used to directly treat a disease process. For example, endothelial growth factors may enhance the biocompatibility of the stent-grafts by accelerating endothelialization.
Biologic molecules other than collagen may also prove to be useful as stent-graft materials. Metal stents that had 3-mm diameters and that were covered with a fibrin film showed promising results in a porcine model, with 90% patency at 28 days and with no giant cell reaction to foreign bodies (10). Graft materials made from a combination of naturally occurring molecules (such as collagen, fibrin, heparin, and growth factors) may someday be designed for use in the treatment of vascular diseases while eliciting little or no adverse reaction to the implant.
Although much is known about metallic endovascular stents and about graft materials that are surgically anastomosed to arteries, little is known about the effect of covering the diseased or normal endothelium with stent-grafts. Ombrellaro et al (31) demonstrated increased graft reendothelialization and decreased intimal thickening with polytetrafluoroethylene-covered Palmaz stent-grafts, compared with surgically anastomosed polytetrafluoroethylene grafts.
The lack of an endothelium on the luminal surface of the grafts has been implicated as a stimulus for the activation of smooth muscle cells, the development of intimal hyperplasia, and the late failure of grafts (32–34). The formation of an endothelium provides a nonthrombogenic and antiinflammatory barrier between the stent-graft and the circulating blood (31,33). Type I collagen has been shown to support rapid endothelial adhesion and growth (35,36). The collagen material used in our study was completely endothelialized at 1 month after implantation; this may be helpful in the maintenance of long-term patency.
Nitinol stents have been shown to be biocompatible (37). The presence of the metal stent may cause turbulence, thrombus formation, or endothelial injury, which could alter the reaction to the adjacent collagen. The lack of complete apposition of the collagen against the endothelium results in areas in which the blood accumulates, clots, and then organizes. This results in the further narrowing of the vascular lumen. An improved design that uses a metal stent to support the entire stent-graft could result in the better apposition of the collagen to the luminal surface of the artery.
Atherosclerotic lesions (especially those in smaller arteries) treated with simple metal stents are prone to subacute thrombosis and restenosis (38,39). This problem has stimulated research in the use of metal stents that are modified with other materials, which include synthetic polymers (10,11), fibrin (10), genetically engineered and cultured endothelial cells (40), radioactive substances (41), and pharmaceuticals (42,43).
The exclusion of the atherosclerotic lesion from direct contact with intraluminal blood by using nonporous stent-grafts has been suggested as a means of reducing the occurrence of stent-related intimal hyperplasia and restenosis (44). However, this theory remains unproved. If stent-grafts are found to be useful in the treatment of atherosclerosis, type I collagen may be a good graft material to use as a scaffold for reendothelialization, healing, barrier formation, and the delivery of drugs to the areas adjacent to the excluded atherosclerotic lesion.Practical application: Our study findings demonstrate that a bovine type I collagen preparation, as a material for endovascular stent-grafts, is biocompatible in rabbits for at least 3 months. Future experiments will test the use of this collagen as a covering for balloon-expandable stent-grafts that will deploy more reliably and that will have greater clinical applicability. After these improvements, experiments will be conducted in animal models of vascular diseases, including atherosclerosis and aneurysm.
|
Footnotes |
|---|
3 St Vincent Mercy Medical Center, Toledo, Ohio.
4 Emory University Hospital, Atlanta, Ga.
H.B.L. and S.T.L. are shareholders in RenGen, which manufactures the collagen stent-graft material used in this study.
Author contributions: Guarantor of integrity of entire study, H.J.C.; study concepts, H.J.C., D.F.K., H.B.L., S.T.L., G.A.H., J.E.D., M.E.J.; study design, H.J.C., D.F.K., H.B.L., S.T.L., G.A.H.; definition of intellectual content, H.J.C., D.F.K., H.B.L., S.T.L., M.E.J.; literature research, H.J.C., D.F.K.; experimental studies, H.J.C., D.F.K., H.B.L., S.T.L., W.F.M., S.B.H., J.K.M.; data acquisition, H.J.C., D.F.K., H.B.L., S.T.L., W.F.M., S.B.H., M.B.L., J.K.M.; manuscript preparation, H.J.C., D.F.K.; manuscript editing, H.J.C., D.F.K., H.B.L., S.T.L.; manuscript review, M.E.J.
|
References |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. R. Whittaker and M. F. Fillinger The Engineering of Endovascular Stent Technology: A Review Vascular and Endovascular Surgery, March 1, 2006; 40(2): 85 - 94. [Abstract] [PDF]
|
|
|
|
 |
|
 J. F.W. Keuren, S. J.H. Wielders, A. Driessen, M. Verhoeven, M. Hendriks, and T. Lindhout Covalently-Bound Heparin Makes Collagen Thromboresistant Arterioscler. Thromb. Vasc. Biol., March 1, 2004; 24(3): 613 - 617. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
