The explosive growth in arterial surgery over the last 45 years has in large part been dependent on the increased use of arterial substitutes. It has been estimated that in the United States alone, over 350,000 synthetic arterial grafts are implanted each year; the number of peripheral autogenous vein grafts exceeds 200,000 per year. This large use of arterial grafts indicates that development of the optimal arterial substitute is of great clinical importance. However, critical review of the present results of arterial grafting leads to the conclusion that the ideal arterial substitute has not yet been developed.
The optimal arterial substitute should (1) be strong, inexpensive, and capable of lasting the life of the patient; (2) be easily and permanently attachable to the host vessel; (3) be biocompatible with the host and have a nonthrombogenic luminal surface; (4) resist infection; (5) be readily available in appropriate sizes; (6) remain patent without subsequent intervention; and (7) have viscoelastic properties similar to those of a normal artery.
An ideal vascular graft should not (1) leak blood or serous fluid with restoration of flow; (2) degenerate chemically or physically with time; (3) incite an abnormal proliferative response from the native vessel or the surrounding tissue; (4) promote thrombus formation or be a source of embolic material; (5) occlude when flexed; or (6) damage blood components. No currently available arterial substitute approaches these requirements; hence the large amount of clinical and basic research devoted to the development and evaluation of vascular grafts.
HISTORICAL ASPECTS
In 1906, Carrel and Guthrie first reported successful implantation of venous autografts into the arterial systems of dogs. They observed that these autografts underwent rapid structural change consisting primarily of a marked thickening of the connective tissue in the adventitia and media. They also noted improved patency when the calibers of the vein and the artery to which it was anastomosed were similar. This was soon followed by clinical use of the popliteal vein for arterial reconstruction after popliteal aneurysm excision by Goyanes in 1906. The first use of a saphenous vein graft in popliteal artery reconstruction after popliteal aneurysm excision in the United States was by Bernheim in 1915.
The first successful arterial allograft was reported by Hoepfner in 1903. Carrel performed a series of experimental arterial autografts and allografts several years later, accompanied by detailed microscopic studies. He found that fresh arterial autografts functioned well and remained microscopically normal during several months of observation. Viable, refrigerated allografts and nonviable, preserved allografts developed progressive wall thickening and hyalinization, depending generally on the type of preservative used and the duration of refrigeration. Nonviable grafts killed by heat, formalin, or glycerin showed rapid degeneration, accompanied by significant host fibrous reaction.
The monumental work of Carrel and Guthrie established the feasibility of arterial and venous autografts and allografts early in the twentieth century. Not until almost 50 years later, however, was widespread clinical application of arterial reconstruction feasible. The pioneering work of Murray on the intraoperative use of heparin, the work of Moniz and dos Santos and colleagues in establishing the technique of arteriography, subsequently combined with the concept of arterial substitutes to initiate the modern era of clinical vascular grafting.
EVALUATION OF ARTERIAL SUBSTITUTES
Patency is the most important end-point in the evaluation of the clinical performance of any arterial substitute. It is critical to distinguish between primary and secondary patency and assisted primary patency. Primary patency is that achieved without any additional graft-directed procedures. Secondary patency refers to grafts that have been maintained patent by one or more additional graft-directed procedures, regardless of whether the graft thrombosed prior to revision. If a later operation involves only the inflow or outflow of the graft and not the anastomoses of the graft or the graft itself, the graft may still be regarded as primarily patent. Assisted primary patency, a disputed term in limited use, refers to secondarily patent grafts that undergo revision prior to actual graft thrombosis.
The concepts of primary, assisted primary, and secondary patency are all important. Primary patency reflects the natural history of individual arterial substitutes. Secondary patency is an indicator of the long-term functional effectiveness of a graft. Assisted primary patency has come to be an indicator of the effectiveness of a clinical follow-up program to detect failing grafts prior to thrombosis. Clearly, however, primary patency is the most important factor in assessing the true overall value of a graft.
Ideally, criteria for patency should be uniformly accepted and clearly stated. It is especially important that life-table analysis be used to display patency. Unfortunately, many reports of arterial substitutes have not employed uniform methods of obtaining and reporting data. Primary and secondary patencies are frequently confused, especially in older publications. However, standards for reports of lower extremity revascularization now exist. 26 Their uniform use will clearly result in more meaningful data in the future, permitting more accurate evaluation of arterial substitutes. In the following sections, the available arterial substitutes, including allografts, xenografts, autografts, and prosthetic grafts, are reviewed.
ALLOGRAFTS
Arterial Allografts
Arterial allografts were the first widely used arterial substitutes. Gross and associates reported the use of viable arterial allografts in patients in 1948. Early results were encouraging, and it was soon recognized that tissue viability was not essential for successful grafting, provided the vessels were properly preserved. This realization, in combination with increasing demand for allografts, led to the establishment of human arterial banks in the late 1940s, freeze-drying being the most popular method of allograft preservation.
Widespread clinical use of arterial allografts in the late 1940s and early 1950s produced a rapid increase in knowledge of the biology and natural history of this type of arterial substitute. Arterial allografts rapidly lose endothelium. A platelet-fibrin coagulum forms on the exposed basement membrane and slowly undergoes fibrous organization. This process begins in anastomotic sites and is frequently incomplete in the central area of the allograft, leaving this area permanently covered with only a fibrin coagulum and prone to ulceration. Allograft walls become less cellular with time. Progressive thinning of the wall, with loss of collagen and fragmentation of elastic fibers, typically occurs after several years. Similar degenerative changes affect both muscular and elastic arterial allografts, but they occur much more rapidly with the former. Thus, allografts of the aorta, composed predominantly of collagen and elastin, have been associated with fewer complications (thrombosis, calcification, aneurysm formation, rupture) and longer graft function than femoral artery allografts, which contain a large component of smooth muscle and elicit a more prominent rejection reaction. However, with the exception of short segment repairs for aortic coarctation, even aortic allografts have disappointing long-term results. They have a high closure rate after only a few years, occasionally accompanied by dilatation and/or rupture.
With the high incidence of complications, arterial allografts have been abandoned in favor of more satisfactory arterial substitutes. Allografts, however, occupy an important place in the history of vascular surgery. The modern era of vascular grafting began with the successful clinical use of arterial allografts by Gross. Oudot and Beaconsfield used an aortic allograft for the first aortic resection and replacement for occlusive disease; DuBost and colleagues used one for the first excision and grafting of an abdominal aortic aneurysm. Studies in animal models suggest that the immunologic rejection process that contributes to the degeneration of these grafts may be modified with low doses of cyclosporine. 27 Such studies raise the intriguing possibility of new applications for these grafts.
Saphenous Vein Allografts
Saphenous vein allografts from human cadavers generally have proved unsatisfactory in clinical practice. Initial encouraging reports were quickly overwhelmed by numerous studies demonstrating a high failure rate in the first postoperative year, as well as late aneurysmal degeneration. 20
Like arterial allografts, venous allografts are normally antigenic and elicit an immunologic rejection response by the host. Microscopic analysis of failed saphenous vein allografts reveals areas of wall necrosis and intimal disruption. 41 Cryopreservation alone does not alter allograft immunogenicity, and the suggestion by some that cryopreservation may enhance vein allograft function has not been confirmed. However, the ability to preserve veins, coupled with advances in immunosuppression, may eventually lead to the establishment of practical antigen-defined allograft vein banks. Currently, however, saphenous vein allografts have no significant clinical application. The only vascular allograft that has enjoyed widespread use is the umbilical vein allograft.
Umbilical Vein Allografts
Umbilical vein allografts have been used primarily for lower extremity revascularizations and were developed as an alternative to autogenous vein. The grafts are prepared from human umbilical cords and are subjected to glutaraldehyde tanning and multiple ethanol extractions and are externally reinforced with a polyester mesh tube. This conduit has a bursting pressure approaching 1000 mm. Hg and is essentially nonantigenic. The largest clinical experience has been accumulated by Dardik and co-workers, who reported primary patency rates of 70% and 50% at 1 and 5 years for femoropopliteal grafts and 50% and 25% at 1 and 5 years for femorotibial grafts.8 These results are distinctly inferior to patency rates for saphenous vein autografts in the same locations. Randomized prospective evaluation of umbilical vein grafts compared with polytetrafluoroethylene (PTFE) grafts (discussed later) indicates comparable patency rates for both grafts when used as below-knee femoropopliteal bypasses. 10
Umbilical vein grafts have several disadvantages that have precluded widespread clinical use. The grafts exhibit degenerative changes over time and are prone to the development of aneurysms. They are technically difficult to implant, and the intima is easily damaged by clamps or attempts at thrombectomy. Infection of umbilical vein conduits requires their removal, and reoperative dissections are often difficult. At present, umbilical vein grafts have no well-defined role in the modern practice of vascular surgery, and they are used with decreasing frequency.
XENOGRAFTS
Unmodified arterial xenografts were used clinically in the early 1950s. These grafts elicited a prominent host immunologic reaction, leading to severe damage to the graft wall. Their use was associated with a high incidence of thrombosis and rupture, and it soon became obvious that unmodified arterial xenografts were not suitable for clinical use.
Rosenberg and associates produced modified xenografts by treating bovine carotid arteries with the proteolytic enzyme ficin, followed by tanning with dialdehyde starch. The result was an almost nonantigenic, collagenous tube that was devoid of smooth muscle and elastic tissue but possessing the same tensile strength as a normal artery. Modified xenografts were used frequently as arterial substitutes from the mid-1960s to mid-1970s but did not produce satisfactory clinical results, particularly in infrainguinal reconstructions. When used in the femoropopliteal position, these grafts were plagued by perioperative thrombosis and exhibited poor long-term patency rates of about 40% 3 to 6 years following implantation. In addition, aneurysms occurred in 3% to 6% of grafts, usually several years after placement. These figures represent aneurysm formation in all implanted grafts. The incidence of aneurysm formation in grafts that remained patent was undoubtedly considerably higher.
Moreover, clinical use of these grafts was associated with an unacceptable infection rate of 3% to 7%, several times that associated with other arterial substitutes. Nevertheless, bovine xenografts have proved to function satisfactorily as hemodialysis access shunts and remain the preferred conduit for this purpose in some centers. They currently have no other well-defined role.
AUTOGRAFTS
Arterial Autografts
The clinical use of arterial autografts was introduced by Wylie in 1965. Proponents of autografts cite their numerous advantages, including retention of viability associated with the maintenance of an intact intrinsic blood supply during grafting, absence of aneurysmal degeneration, resistance to infection, preservation of normal flexibility at points of joint motion, and possession of growth potential when used in pediatric patients. The obvious disadvantage is a lack of availability, including inadequate length of arteries for most potential applications.
The innermost 300 to 500 m. of the arterial wall is nourished by luminal diffusion.3 The outer layers of the arterial wall are nourished by a complex vasa vasorum system derived primarily from the most proximal portions of arterial side branches. If the proximal portions of side branches are excised or damaged during preparation of the autograft, spotty arterial wall necrosis may develop in that portion of the arterial wall more than 500 to 600 m. from the lumen.
The clinical use of arterial autografts is restricted primarily to coronary artery bypass grafting, renal artery bypass in children, and arterial substitutes in infected surgical fields. Radial arteries have been used as free grafts in coronary bypass surgery, but their use has been largely unsuccessful due to early graft closure associated with florid intimal proliferation. Currently, however, the internal mammary artery is frequently used as a conduit in coronary bypass surgery and provides patency rates that are clearly superior to those of saphenous vein grafts.
There is general agreement that children requiring renal artery reconstruction should have arterial autografts. The internal iliac artery has proved to be a suitable conduit, and its use obviates the high incidence of late aneurysmal graft dilatation in this age group when saphenous vein grafts are used. Iliac artery autografts have also proved to be remarkably effective in a limited number of adult patients undergoing renal vascular surgery. In a primarily adult population, Stoney and associates reported only two arterial autograft occlusions in 86 patients undergoing renal vascular surgery, with follow-up extending 16 years. 34 Use of the internal iliac artery is, however, limited in the adult population by its frequent involvement with advanced atherosclerosis. The external iliac artery may also be used as an arterial autograft, but it generally requires prosthetic replacement of the autograft donor site.
Arterial autografts are occasionally used to bridge short arterial defects in infected or contaminated surgical fields that preclude the use of synthetic grafts. Atherosclerotic arteries may be endarterectomized and used as either bypass conduits or, more frequently, patch grafts. These technical adjuncts are employed primarily in the management of traumatic wounds or infected prosthetic grafts. Unfortunately, the use of endarterectomized arteries as autografts has been associated with a disturbing incidence of fibrointimal hyperplasia leading to graft failure.
Venous Autografts
The routine clinical use of venous autografts began with the clinical report by Kunlin in 1949. Since that time, venous autografts have proved to be the most successful and clearly the most clinically important small-caliber arterial substitutes. They are the preferred graft for infrainguinal arterial reconstruction. The greater saphenous vein, lesser saphenous vein, cephalic and brachial veins, and superficial femoral and internal jugular veins have all been used as bypass conduits.
Greater Saphenous Vein. Greater saphenous vein autografts are by far the most widely used autogenous vascular graft in modern vascular surgery and are currently the standard with which all other small-caliber arterial substitutes are compared. Over 200,000 peripheral vascular operations are performed in the United States annually.
The normal greater saphenous vein averages 70 to 80 cm. in adult men. It begins at the medial malleolus at the junction of the medial marginal and internal malleolar veins. This vein is quite superficial in the leg but lies close to the deep fascia in the thigh. It is a single vessel in the thigh in 75% of patients; contains 8 to 12 bicuspid valves, mainly below the knee; and averages 5.5 mm. in diameter. The luminal surface consists of a monolayer of endothelial cells. The media is composed of an inner layer of longitudinally arranged smooth muscle cells and an outer layer of circumferentially oriented smooth muscle cells. The outer adventitial layer is composed of a loose mixture of collagen and elastic tissue.
The saphenous vein has been used as a replacement for small and medium-sized arteries in all parts of the body, with most being used for coronary artery bypass grafts and for lower extremity bypass of occluded superficial femoral, popliteal, and tibial arteries. Other less frequent uses include upper extremity bypass, as well as mesenteric and renal artery bypass. The coronary, mesenteric, and renal results are described in other sections and are generally excellent.
Lower extremity bypasses using autogenous saphenous vein may be performed using one of two basic techniques. An appropriate length of vein may be removed from either the ipsilateral or the contralateral lower extremity; it is then reversed in direction to permit arterial flow in the direction of the venous valves and sutured in place, usually in an end-to-side configuration. Alternatively, an intact ipsilateral vein of adequate quality may largely be left in its anatomic position and the valves destroyed using one of a variety of intraluminal devices, followed by similar proximal and distal arterial anastomoses. This technique has been termed in situ saphenous vein bypass. Both techniques are currently widely employed and generally produce similar results.
The primary patency of reversed femoropopliteal saphenous vein autografts ranges from 80% to 90% at 1 year, 55% to 86% at 5 years, and 38% to 46% at 10 years (Fig. 50–8 Fig. 50–8). 38 Reversed saphenous vein grafting to tibial arteries produces patency rates that are about 10% to 15% lower than those for femoropopliteal grafting at all time intervals. 38 Exhaustive analyses of variables affecting femoropopliteal patency indicate that patency is higher when bypass surgery is performed for claudication rather than limb salvage, and it is generally higher with a widely patent popliteal and tibial artery outflow tract. A patent outflow tract, however, is not an absolute requirement for long-term patency. Mannick and co-workers reported an intermediate-term patency rate of 65% when bypassing to an isolated popliteal segment without demonstrated angiographic patency of the popliteal artery trifurcation vessels. 19 A patient's continued cigarette smoking and the use of a small vein (less than 4 mm. after gentle distention) both decrease long-term patency. Curiously, vein graft patency appears to be slightly higher in diabetic patients. 38 The role of antiplatelet drugs in enhancing vein graft patency is unproven, although considerable anecdotal evidence supports their use.
Large series of in situ vein bypass initially suggested that the technique resulted in superior patency rates compared with the reversed vein technique. 18 Modern series of reversed vein bypasses, however, have demonstrated similar or superior patency rates to in situ bypasses in grafts to the popliteal and tibial arteries. 38 Clearly, reversed vein grafting is applicable to larger numbers of patients, because many patients do not have an intact ipsilateral saphenous vein, which is mandatory for an in situ bypass.
Alternative Venous Autografts. Venous conduits other than the greater saphenous vein are now recognized as being suitable for lower extremity bypass.
As many as 20% to 30% of patients in need of leg bypass do not possess greater saphenous veins adequate for arterial grafting. 37 The vein is anatomically too small in 5% to 10% of patients and is unavailable or unusable in another 10% to 20% of patients because of prior removal, previous thrombosis, or varicosities. In such patients, alternative veins have been used for arterial grafting, including the lesser saphenous, basilic, cephalic, and superficial femoral. Various venous segments may be joined with venovenostomies to achieve an autogenous conduit of adequate length. 37
Although the long-term effectiveness of these alternative conduits has been questioned, recent data suggest that they can serve as acceptable arterial substitutes. Harris and his associates reported 3-year patency rates of 82% and 65% using cephalic vein bypasses to the below-knee popliteal and tibial arteries, respectively. 14 The superficial femoral vein has been used as a femoropopliteal bypass conduit with a primary patency rate of 82% at 3 years. 28 Some report surprisingly little postoperative morbidity with the use of superficial femoral veins as bypass grafts, but most surgeons are reluctant to use them as a first-choice autogenous arterial substitute because of concerns about postoperative edema. Other investigators have also found alternative vein sources to be satisfactory arterial substitutes in the lower extremity. 37
Pathology of Venous Autografts. Early postoperative vein graft failure can usually be attributed to technical flaws in the performance of the operation or the presence of an unrecognized hypercoagulable state. Late failures may follow progression of atherosclerotic disease above or below the vein graft. It is quite clear, however, that the vein graft itself is subject to a number of pathologic alterations that may contribute to occlusion of the bypass.
Carrel first noted that veins implanted into the arterial system that remain patent inevitably undergo significant thickening. This process, which has been erroneously termed “arterialization” of vein grafts, produces medial and subintimal fibrous hyperplasia, often in combination with fibrin deposition on the intimal surface. 31 The result is a variable thickening of the vein wall that may be minimal and remarkably localized or may involve the entire vein graft, with resultant diffuse obliteration of the lumen.
Szilagyi and colleagues, in a landmark review of the clinical outcome of peripheral arterial saphenous vein grafts, found that marked fibrointimal hyperplasia occurred in 8% of such grafts. 35 Interestingly, this process has been implicated as the cause of failure in 15% to 30% of aortocoronary grafts occluding during the first year 32 and is the probable cause of at least two thirds of all infrainguinal vein bypass graft failures. An intense search continues for the cause of fibrointimal hyperplasia and ways to prevent it.
Fibrointimal hyperplasia follows the conversion of normally quiescent myointimal and medial smooth muscle cells into actively proliferating secretory myofibroblasts. A number of factors appear to be important in stimulating this proliferative process. The predominant current theory regards fibrointimal hyperplasia as a response to vein injury occurring during and following grafting. Without doubt, veins are mechanically injured during removal and storage preparatory to arterial grafting. In addition, the vein wall is rendered ischemic during its dissection because of disruption of the vasa vasorum. Electron micrographs after vein harvest frequently show large areas of endothelial denudation and medial injury. Although there is agreement on the importance of gentle vein harvest techniques and the avoidance of excessive hydrostatic venous distention, there is little agreement on the optimal medium in which to store veins for reversed bypasses between harvest and arterial grafting. Some have found minimal endothelial disruption with storage in chilled autogenous blood. Others have recommended placing the vein in tissue culture medium at 37º C. with papaverine. Many surgeons simply leave the vein in a heparinized saline solution for the short time between vein excision and implantation.
In addition to the vein harvest itself, other mechanical factors may be important in inciting fibrointimal hyperplasia. These include increases in shear forces and venous wall stress induced by arterial pressure and compliance mismatch at anastomotic sites. The potential importance of the latter is suggested by the observation that failed saphenous vein grafts frequently have prominent fibrointimal hyperplasia at the proximal or distal anastomosis. Recent animal studies, however, suggest that although compliance mismatch may contribute to graft thrombosis, there does not appear to be any significant difference in anastomotic fibrointimal hyperplasia in compliant versus noncompliant grafts. 22
Endothelial damage appears to be the final common pathway in the production of fibrointimal hyperplasia. Whereas in the past the endothelium was regarded as merely an inert nonthrombogenic surface lining, it is now clear that these cells are very active biochemical factories capable of responding to injury and producing a variety of substances involved in the regulation of vascular wall function.
Platelets, endothelium, smooth muscle cells, and macrophages can all produce similar growth factor proteins that can induce smooth muscle cell proliferation. Endothelial cells also produce growth-inhibiting factors (heparan sulfate), which suggests that endothelial cells may be able to modulate various mitogens capable of producing fibrointimal hyperplasia. The mechanisms by which alterations in cellular interactions and growth factor production combine to produce fibrointimal hyperplasia leading to a failing or failed vascular graft are currently the focus of intense investigation and have important clinical implications.
A number of other pathologic changes have been noted in vein grafts. Clamp trauma may produce localized stenosis associated with transmural fibrosis. This has been reported in 4% of lower extremity vein grafts. 35 About 9% of vein grafts develop localized stenosis following fibrosis of the venous valves or suture narrowing caused by improper suture ligation of venous side branches. Significant vein graft atherosclerosis occurs after variable periods in approximately 7% of aortocoronary grafts and 15% of lower extremity grafts, causing localized stenosis in about 7% of the latter. 36 Atherosclerotic venous aneurysms develop in 3% to 8% of lower extremity vein grafts. 36 The incidence of nonatherosclerotic vein graft dilation or aneurysm formation varies significantly with the location of the graft and the age of the patient. Stanley and colleagues noted that although one third of aortorenal vein grafts become ectatic, actual aneurysmal degeneration occurs in only 1.5% of adults. In children, however, the incidence of aneurysmal degeneration with similar grafts is 20%, a finding that has led to the current preference for arterial autografts in children requiring renal artery reconstruction. 33
About one third of vein grafts eventually develop recognizable structural defects. 36 The majority of these defects are stenotic lesions and develop within 1 year of graft implantation. If not corrected, many lead to failure of the graft. In the past, this has led some surgeons to recommend routine postoperative angiography at 1- to 2-year intervals. 36 Currently, however, it appears that noninvasive determination of blood flow velocity within the graft by means of duplex ultrasonography is quite accurate in predicting subsequent graft thrombosis. Patients with graft flow velocities in the midportion of the graft below 45 cm. per second or localized high peak systolic velocities (greater than 200 cm. per second) within the graft or at its proximal or distal anastomoses should undergo angiography to locate a potentially correctable lesion. 1
PROSTHETIC GRAFTS
Textile Grafts
The development of prosthetic arterial grafts was stimulated by the observation by Voorhees in the early 1950s that silk threads in the canine vascular system became covered by a glistening endothelium-like cellular coating. The hypothesis was then proposed that a fine mesh fabric would result in similar healing and thus function as a satisfactory arterial substitute.
Voorhees and associates subsequently described successful replacement of arteries in animals with a porous textile graft made from the nylon derivative Vinyon-N. Two years later, the same graft was successfully implanted in humans. The field of prosthetic vascular grafts has since achieved enormous clinical and laboratory importance.
Composition and Fabrication. Both the material and its method of fabrication are important in the manufacturing of prosthetic arterial substitutes. Materials such as nylon, Orlon, Ivalon, and Marlex have all proved disappointing, primarily because of loss of tensile strength and kinking. The only textile materials thus far that have proved to function satisfactorily are Dacron and Teflon, neither of which loses significant tensile strength even after many years of implantation.
Dacron and Teflon grafts are manufactured by weaving or knitting multifilament texturized yarns (Figs. 1 and 2). Each process has advantages and disadvantages. Woven grafts must be tightly interlaced to prevent slippage and fraying of the yarn. This compact structure of the graft produces small interstices and low porosity. These grafts leak minimally at the time of implantation but are somewhat stiff and difficult to handle. In addition, the tight configuration of the graft theoretically reduces the potential for the development of a living neointima by connective tissue ingrowth through the graft interstices.
Knitted grafts are softer and more compliant than woven grafts, and the knit can be varied. The looser the knit, the more elastic and porous the graft. They have been widely used in vascular surgical operations below the diaphragm because of their excellent handling characteristics, including softness and lack of fraying at cut ends.
Knitted grafts are quite porous, between 1200 and 1900 ml. per cm. per minute, and must be preclotted prior to implantation. (Porosity for graft applications is defined as the amount of water that will pass through 1 sq. cm. of graft wall per minute under a hydrostatic driving pressure of 120 mm. Hg.) A sample of the patient's own blood is forced repeatedly through the graft interstices. This causes platelet-fibrin deposition in the interstices, which renders the graft temporarily impervious to blood. After implantation, this platelet-fibrin material is slowly replaced by fibrous ingrowth from the host. Porous grafts must be used with great caution in patients with platelet or coagulation defects. Under these circumstances, the necessary initial coagulum may never properly form, and the patient may bleed excessively thought the graft interstices. A tightly woven graft is preferred in this setting. Woven grafts are also generally preferred in repairs of the thoracic aorta to limit hemorrhage through the graft interstices, especially in operations requiring full heparinization and cardiopulmonary bypass.
Innovative manufacturing modifications have been superimposed on the basic concepts of knitted and woven textile grafts. Velour surfaces can be added to the inside, outside, or both sides of knitted or woven grafts. Velour surfaces have loops of yarn extending almost perpendicular to the fabric surface (Fig. 3). Various porosities and thicknesses are possible. Velour surfaces improve the handling characteristics of woven grafts and provide a scaffold for fibroblast ingrowth, leading to firm graft adherence to surrounding tissue. The velour concept is widely accepted, and a large percentage of textile vascular grafts in current use have a velour surface.
Most textile grafts in clinical use are also crimped to impart greater flexibility without kinking. Although widely employed, this process has several potential disadvantages. Crimping diminishes luminal diameter, increases the thickness of the graft material, and may also cause deposition of thrombogenic fibrin in the crimped areas. 17 It is possible to manufacture externally supported grafts that avoid kinking with angulation and yield results equal or superior to those obtained with crimped grafts. These grafts have an incompressible large fiber wound around and adherent to the external surface.
Clinical Applications of Textile Grafts. The knitted Dacron graft has been the most frequently used prosthetic arterial graft during the past 30 years. Woven Dacron grafts have traditionally been used primarily in those settings in which interstitial bleeding would present major problems, but the addition of velour surfaces has prompted widespread application of woven grafts. Textile-fabricated Teflon grafts are presently used infrequently, but they generally appear to function satisfactorily, especially in large artery applications.
Textile grafts function most satisfactorily when used for arterial replacement proximal to the inguinal ligament (Figs. 4 and 5). Five- and 10-year patencies have been reported as high as 91% and 66%, respectively, for aortofemoral bypass. 5, 24 Axillofemoral bypass patency has been in the range of 75% to 77% at 5 years, and femorofemoral bypass patency at 5 years has similarly been 75% to 80%. 4, 13, 16 Textile grafts usually produce patency results distinctly inferior to those of saphenous vein bypass below the inguinal ligament. In the most favorable report, externally supported Dacron bypass to the above-knee popliteal artery produced a 5-year patency of 70%. 11 Textile grafts produce low patency results when used to bypass to arteries distal to the popliteal artery; therefore, their clinical use in this setting is not recommended.
Polytetrafluoroethylene (PTFE) Grafts
PTFE is a fluorocarbon polymer formed into sheets by a unique paste extrusion process. Extruded PTFE is not a textile but rather a semi-inert polymer consisting of solid nodes of PTFE with interconnecting small fibrils. The intranodal distance can be varied in the manufacturing process and is about 40 µ for grafts in clinical use (Fig. 6). Although greater intranodal distances permit greater tissue ingrowth and, therefore, theoretically greater graft healing and increased patency, this has not proved to be the case in human trials.
PTFE grafts have a highly electronegative surface charge and are thus hydrophobic and resistant to thrombosis. The grafts are coated with an outer wrap of PTFE to avoid aneurysm formation, which occurred frequently with early clinical use of unwrapped grafts. PTFE grafts are available with external ring supports to avoid compression in subcutaneous locations and kinking with angulation. Wall thickness may be varied, with thin-walled grafts preferred for infrainguinal bypasses. Thick-walled grafts function well as hemodialysis shunts and, in most centers, have replaced bovine grafts as the preferred material for hemodialysis access when creation of a native fistula is not possible.
Clinical Applications of PTFE Grafts. PTFE grafts are available in a wide variety of sizes and configurations suitable for almost any arterial reconstructive procedure. They have been used most widely for construction of extra-anatomic bypasses and as a substitute for autogenous vein in infrainguinal bypasses.
Externally supported PTFE axillofemoral grafts provide 5-year patency rates of about 70%. 13 PTFE grafts in the axillofemoral position have been associated with occasional early postoperative disruption of the axillary artery anastomosis. 39 In some cases, the sutures have pulled through the graft. It is postulated that because PTFE grafts are formed from a continuous extruded polymer, penetration of PTFE grafts by needles may cause alteration of the polymeric structure and subsequent graft disruption when the anastomosis is stressed by shoulder elevation or arm abduction. 39 Positional operative modifications of the PTFE graft–axillary artery anastomosis have greatly reduced the incidence of this complication.
For infrainguinal bypass, initial reports of patency rates similar to those of saphenous vein grafts have been modified by larger series with more extended follow-up. 21 Cumulative patency of PTFE grafts used in the above-knee femoropopliteal position is about 75% at 1 year and 55% at 5 years.
These results, at least in the short term, are sufficiently similar to those reported in certain series of saphenous vein grafts to prompt some to conclude that PTFE grafts should serve as the initial conduit for above-knee femoropopliteal bypass, especially in patients with limited life expectancy. PTFE grafts have performed poorly, however, in comparison with saphenous vein when used as bypass conduits to the below-knee popliteal and tibial arteries or in situations in which there is poor distal runoff. Although some have reported 50% to 60% intermediate patency of PTFE grafts performed for limb salvage, others have concluded that when such grafts must be anastomosed to tibial arteries, the results are so poor that primary amputation should be strongly considered. 9 Despite these considerations, PTFE grafts are widely employed as the prosthetic arterial substitute of choice for infrainguinal bypass when autogenous vein is not available. Patency of PTFE grafts may be extended through the use of routine warfarin anticoagulation. Although PTFE grafts are easily thrombectomized, it now appears that thrombosis of established infrainguinal PTFE grafts is best treated by placement of a new conduit.
Prosthetic Graft Healing
Shortly after prosthetic graft implantation, a thin layer of fibrin is deposited on the luminal surface. In grafts with high flow, the thickness stabilizes at about 1 mm. and is well tolerated. In a low-flow environment, however, the fibrin layer frequently continues to increase in thickness, proceeding to luminal occlusion.
In the experimental animal, the fibrin layer is progressively organized, resulting in the development of a lining consisting of fibroblasts, myofibroblasts, and fibrocollagenous connective tissue. The graft lining then becomes connected to the perigraft tissue by ingrowth through the interstices of the fabric. Prosthetic graft healing refers to the development of a living neointima inside the graft, connected to an external fibrous capsule around the graft by means of connective tissue ingrowth through graft interstices. Well-healed grafts offer the possibility of decreased thrombogenicity and increased resistance to infection.
With respect to the The extent of graft healing is related to both the host species and the graft itself.graft, the quality of healing reflects both the porosity of the graft and the thickness of the graft wall. Healing via tissue ingrowth is favored by increased porosity and decreased wall thickness. Healing may also occur at the graft margins via endothelial ingrowth from the host vessels. Healing by this mechanism is limited to the few centimeters adjacent to the anastomosis. Circulating blood cells, probably stem cells, also appear to aid in the healing of both Dacron and PTFE grafts. 29
Complete graft healing routinely occurs in animal models when thin, porous grafts are used. However, such grafts, when implanted in patients, never develop a complete living neointima. An important clinical study by Berger and associates in 1972 first clearly demonstrated that prosthetic grafts in humans do not develop a living neointimal lining but rather maintain a permanent lining composed primarily of compacted fibrin. 2 These investigators later described detailed studies of prosthetic grafts removed from 64 patients months to years after implantation. The results showed that humans have a limited ability to organize fibrin deposits on the luminal surface of grafts, in sharp contrast to experimental animals. A small zone of luminal healing occurs adjacent to suture lines, but otherwise the graft is typically lined with compacted fibrin even years after implantation. Only two of the 64 grafts showed significant luminal healing, and even then endothelial cells could not be identified. Although this work involved primarily fabric grafts, others suggest a similar situation with PTFE grafts. 7
The concept of limited graft healing in humans is supported by platelet survival and localization studies. Several investigators have correlated platelet survival and platelet-graft deposition with healing of the graft lumen. In the totally healed grafts of animals, both platelet survival and platelet-graft deposition return to normal.
Harker and colleagues have shown that humans with aortic prostheses have shortened platelet survival for 9 months after graft implantation but have increased platelet deposition on the graft for at least 120 months. 12 Thus, all evidence clearly indicates that a large majority of prosthetic arterial grafts in man, unlike in experimental animals, never develop an organized internal lining and remain permanently lined with compacted fibrin. The relevance to man of animal-derived optimal graft design characteristics must be regarded as minimal.
Complications of Prosthetic Grafting
The most frequently observed prosthetic graft complications include anastomotic neointimal hyperplasia, graft infection, graft failure caused by fiber disruption or stretching, perigraft seromas, and development of anastomotic false aneurysms.
Neointimal Hyperplasia. This process is similar to that described for saphenous vein grafts and is present at both proximal and distal anastomoses. Clinically, the distal anastomotic process appears to be more significant and is frequently implicated as a cause of prosthetic graft failure. Both textile and PTFE grafts are affected. The aforementioned etiologic factors for neointimal hyperplasia associated with saphenous vein grafts also apply to prosthetic grafts. As with saphenous vein grafts, antiplatelet drugs are frequently prescribed for patients with prosthetic grafts in an effort to reduce the magnitude of anastomotic fibroplasia. Use of antiplatelet drugs, specifically aspirin, is clearly indicated in patients with peripheral vascular disease. These agents have been shown to decrease significantly the overall cardiovascular-associated morbidity and mortality.
Graft Infection. Infection is one of the most feared complications of prosthetic grafting. Staphylococcus aureus, Staphylococcus epidermidis, and Escherichia coli are the most frequently isolated organisms. The incidence is about 1.5% to 2.5%; it is slightly lower when the graft is completely intra-abdominal and higher when a groin anastomosis is present. Aortic graft infections may be associated with an operative mortality of 10% to 25% and an amputation rate of 15% to 20%. 43 Even infrainguinal graft infections are associated with high mortality and amputation rates. Graft infection may be decreased by careful patient selection, meticulous preparation of sites for surgical incision, and preoperative use of prophylactic antibiotics, usually a cephalosporin. Temporary bonding of antibiotics to prosthetic grafts is possible and is under evaluation in animal models. 30
In a large majority of cases, an infected arterial graft must be removed to control infection. Anecdotal reports of successful treatment of graft infection by local drainage and antibiotic irrigation without graft removal are the exception. The basic principle of infected graft treatment remains graft excision and revascularization through a clean field, usually in an extra-anatomic configuration. It appears that both mortality and amputation rates can be improved if the revascularization is performed first, followed by removal of the infected prosthetic graft. Staging the operation in this manner lessens the complications of prolonged distal ischemia. On occasion, revascularization through the contaminated field may be successful if entirely autogenous tissue is used.
Graft Failure. Two distinct types of Dacron graft failure have been described. The first consists of gradual, diffuse graft dilatation, observed frequently with the ultralightweight knitted Dacron in widespread use in the early 1970s. The dilatation was caused by expansion of the knit rather than elongation or weakening of individual fibers. These diffusely dilated grafts frequently caused no trouble but were occasionally associated with delayed graft rupture or diffuse interstitial bleeding. The ultralightweight method of fabrication has been abandoned by all manufacturers.
The second type of graft failure follows specific defects, such as a dropped stitch in the manufacturing process, or fiber degeneration. This type of defect usually causes localized holes and leaks, with the potential for false aneurysm formation (Fig. 7).
Anastomotic False Aneurysms. An anastomotic aneurysm follows a partial or total separation of the prosthetic graft from the host artery. These are termed false aneurysms, because the blood is contained by the nonelastic surrounding fibrous capsule rather than the true vessel wall. The natural history of an anastomotic aneurysm is progressive expansion, with an eventual serious complication of rupture, thrombosis, or embolism. The incidence of false aneurysm with prosthetic grafting is estimated to be at least 3%. Both Dacron and PTFE grafts are affected.
Because a prosthetic arterial anastomosis is forever dependent upon the anastomotic suture line, many false aneurysms in the past followed degeneration and fragmentation of the silk sutures that were used for construction of the anastomosis. Since the recognition of this problem, silk sutures are no longer used for arterial anastomoses. The exact cause of currently encountered false aneurysms is unknown. Frequently, the suture line and sutures are found intact, with the aneurysm occurring through a tear in the host artery adjacent to the suture line. Obvious clinical infection can be implicated in only a small percentage of cases. With rigorous culture techniques, however, infectious organisms can be demonstrated in a large number of false aneurysms. 40 Whether these subclinical infections are the actual cause of the false aneurysms or merely incidental findings is unknown. Other suggested etiologies of false aneurysms include shallow suture placement or arterial degeneration associated with hypertension. Several interesting studies have suggested that anastomotic aneurysms may follow graft-artery compliance mismatches accentuated by graft dilation. 6 Placement of too short a graft with excessive anastomotic tension has also been implicated. With rare exception, anastomotic aneurysms should be repaired when discovered. The usual repair consists of a graft-artery reanastomosis, frequently with the insertion of a short additional piece of graft material.
Perigraft Seroma. A perigraft seroma is a sterile collection of clear fluid within a nonsecretory pseudomembrane surrounding a prosthetic vascular graft. It usually occurs in association with extra-anatomic bypasses, that is, an axillofemoral or femorofemoral graft. The incidence is probably around 2%, occurring with both Dacron and PTFE grafts. The etiology is unknown but may relate to relative fibroblast inhibition, with failure of graft incorporation into surrounding tissues. Treatment has included observation alone, multiple aspirations, graft removal, injection of perigraft sclerosing agents, excision of the fibrous pseudocapsule, or intraperitoneal drainage.
Graft Modifications and Experimental Arterial Substitutes
A number of modifications of knitted Dacron grafts have been proposed to eliminate the need for preclotting by rendering the grafts temporarily impervious to blood while maintaining desirable handling properties.
Cooley and associates described soaking knitted Dacron grafts in the patient's own plasma, followed by steam autoclaving for 3 to 5 minutes.
Other investigators have used albumin to temporarily coat knitted grafts. These grafts require special packaging and must be rehydrated prior to implantation, but they do handle somewhat better than standard low-porosity woven arterial substitutes.
Dacron grafts may also be impregnated with bovine Type I collagen. 25 The grafts are leakproof and handle essentially the same as standard knitted grafts. Cross-linking of the collagen during the manufacturing process seems to eliminate the normal thrombogenicity of collagen, and the grafts appear to be no more prone to thrombosis than a standard preclotted, knitted prosthesis. In an effort to reduce intraoperative blood loss, collagen-impregnated knitted Dacron grafts have become the graft of choice for many surgeons performing aortic reconstructions.
Bovine carotid xenografts have been modified with a combination of amino acid carboxylation and glutaraldehyde tanning. The negative luminal electrical charge imparted by this process supposedly makes the grafts resistant to thrombosis but should be regarded as experimental and of unproven clinical efficacy. The same applies to autogenous fibrocollagenous tubes formed over a silicone mandril, pyrolytic carbon-coated grafts, polyurethane grafts of various configurations, and Dacron-PTFE combinations.
Many studies have been published over the last 10 to 15 years concerning endothelial seeding of small-caliber synthetic vascular prostheses. It was hoped that the presence of a living neointima composed of viable endothelial cells would result in improved patency of prosthetic infrainguinal bypasses.
The research is complex, including problems in harvesting viable endothelial cells, optimizing methods of seeding and cellular attachment to the graft, tracking of cells from harvesting to healing, and documenting normal function of seeded endothelial cells. Clinical trials of endothelialized PTFE grafts are beginning to be reported, with mixed but generally negative results. Zilla and colleagues randomized 49 patients with no saphenous veins who required femoropopliteal bypass to either endothelial seeded or standard PTFE grafts. After 32 months, life-table analysis revealed a patency of 85% for the endothelialized grafts and 55% for the control PTFE grafts. 44 Herring and colleagues compared 66 endothelial seeded femoropopliteal PTFE grafts with 53 autogenous vein femoropopliteal grafts. At 30 months, vein graft patency was 92%, and the patency of the seeded PTFE grafts was 38%. 15
Another related and rapidly developing field involves the implantation of vascular grafts with genetically modified endothelial cells. 42 Researchers are exploring the possibility of using the vascular endothelial cell as a target for gene replacement therapy. The luminal location of the endothelial cell theoretically makes it ideal for the delivery of circulating gene products. In addition, this technology may prove important in prolonging graft patency by the delivery of gene products acting in a paracrine fashion to inhibit anastomotic intimal hyperplasia.
Currently, there is intense worldwide interest in the development of transluminally placed prosthetic arterial substitutes for the treatment of aneurysms. Dacron or PTFE grafts are mounted on expandable metallic stents (Fig. 8). An artery distal to the site to be grafted is surgically exposed. A coaxial delivery system under fluoroscopic guidance is then used to transluminally deliver the endoluminal stent graft to the desired location, where it either self-expands or is expanded with an angioplasty balloon, to exclude the flow stream from the aneurysm sac. The grafts are held in place by the outward pressure of the stent itself and/or by hooks attached to the proximal end of the stent portion of the prosthesis that engages the artery wall.
There are a number of potential problems with stent grafts, including the large size of the delivery systems currently required to place the grafts, the need for suitable anchoring sites, graft migration, arterial perforation, possible embolization of the luminal thrombus that is inevitably present within aneurysms, perigraft leakage such that the aneurysm is not completely excluded from the flow stream, and graft kinkage and axial twists in the limbs of bifurcated grafts. Despite these difficulties, a number of patients worldwide have had placement of endoluminal stent grafts for arterial disease. 23 The ultimate role of stent grafts as arterial substitutes awaits completion of careful clinical trials and further advances in the design of the grafts and their delivery systems.
OVERVIEW
Arterial grafting is of critical importance in modern vascular surgery. Available evidence clearly indicates that large artery bypass is best accomplished with textile-fabricated Dacron grafts or with PTFE grafts. For small artery bypass below the inguinal ligament, no arterial prosthesis has matched the performance and patency of autogenous saphenous vein grafts.
To date, no arterial prosthesis 4 mm. or smaller in diameter has produced satisfactory clinical patency. Various prosthetic grafts to the above-knee popliteal artery have approached intermediate-term vein bypass patency, although none is as good as vein. No prosthetic graft has approached saphenous vein results for bypass to the below-knee popliteal artery or to tibial arteries.
Research in graft design and fabrication is currently directed toward the development of satisfactory prosthetic grafts for small artery bypass and endoluminally placed arterial substitutes. Most investigators believe that a satisfactory small artery prosthesis requires luminal healing and compliance characteristics that are considerably superior to those of grafts in present use, accompanied by improved understanding of the processes of atherosclerosis, thrombosis, and arterial healing.
SELECTED REFERENCES
Bandyk, D. F., Cato, R. F., and Towne, J. B.: A low flow velocity predicts failure of femoropopliteal and femorotibial bypass grafts. Surgery, 98:799, 1985.
This landmark paper paved the way for the use of duplex scanning in the postoperative evaluation of autogenous vein grafts. The use of noninvasive techniques has greatly facilitated understanding of the natural history of autogenous vein grafts.
Berger, K., Sauvage, L. R., Rao, A. M., and Wood, S. J.: Healing of arterial protheses in man: Its incompleteness. Ann. Surg., 175:113, 1972.
This article reports detailed pathologic observations on arterial protheses removed from patients. It remains of pivotal importance in emphasizing the marked differences between man and animal in the healing of prosthetic arterial grafts.
Parodi, J. C., Palmaz, J. C., and Barone, H. D.: Transfemoral intraluminal graft implantation for abdominal aortic aneurysms. Ann. Vasc. Surg., 5:491, 1991.
With this publication, Dr. Parodi and colleagues launched an area of investigation that will dominate the field of arterial substitutes for a number of years.
Sawyer, P. N. (Ed.): Modern Vascular Grafts. New York, McGraw-Hill, 1987.
This multiauthored text contains literature reviews of various grafts by noted authorities.
Szilagyi, D. E., Elliott, J. P., Hageman, J. H., Smith, R. F., and Dal'Olmo, C. A.: Biologic fate of autogenous vein implants and arterial substitutes. Ann. Surg., 178:232, 1973.
This excellent reference clearly documents, with detailed clinical, angiographic, and pathologic follow-up, the long-term performance of autogenous vein grafts in man.
REFERENCES
1. Bandyk, D. F., Cato, R. F., and Towne, J. B.: A low flow velocity predicts failure of femoropopliteal and femorotibial bypass grafts. Surgery, 98:799, 1985.
2. Berger, K., Sauvage, L. R., Rao, A. M., and Wood, S. J.: Healing of arterial prostheses in man: Its incompleteness. Ann. Surg., 175:118, 1972.
3. Berger, K., Sauvage, L. R., Wood, S. J., and Sameh, A. A.: Endarterectomy and other surgical injuries to cardiovascular walls. Pacific Med. Surg., 75:367, 1967.
4. Brief, D. K., Brener, B., and Alpert, J.: Crossover femoro-femoral grafts followed up 5 years or more: An analysis. Arch. Surg., 110:1294, 1975.
5. Brewster, D. C., and Darling, R. C.: Optimal methods of aortoiliac reconstruction. Surgery, 84:739, 1978.
6. Clagett, G. P., Salander, J. M., and Eddleman, W. L.: Dilatation of knitted Dacron aortic prostheses and anastomotic false aneurysms: Etiologic considerations. Surgery, 93:9, 1983.
7. Clowes, A. W., Kirkman, T. R., and Reidy, M. A.: Mechanisms of arterial graft healing: Rapid transmural capillary ingrowth provides a source of intimal endothelium and smooth muscle in porous PTFE prostheses. Am. J. Pathol., 123:220, 1986.
8. Dardik, H., Miller, N., Dardik, A., Ibrahim, I. M., Saussman, B., Berry, S. M., Wolodiger, F., Kahn, M., and Dardik, I.: A decade of experience with the glutaraldehyde-tanned human umbilical cord vein graft for revascularization of the lower limb. J. Vasc. Surg., 7:336, 1988.
9. Dennis, J. W., Littooy, F. N., Greisler, H. P., and Baker, W. H.: Secondary vascular procedures with polytetrafluoroethylene grafts for lower extremity ischemia in a male veteran population. J. Vasc. Surg., 8:137, 1988.
10. Eickhoff, J. H., Broome, A., Ericsson, B. F., Hansen, H. J. B., Kordt, K. F., Mouritzen, C., Kvernebo, K., Norgren, L., Rostad, H., and Trippestad, A.: Four years' results of a prospective, randomized clinical trial comparing polytetrafluoroethylene and modified human umbilical vein for below-knee femoropopliteal bypass. J. Vasc. Surg., 6:506, 1987.
11. El-Massry, S., Saad, E., Sauvage, L., Zammit, M., Smith, J., Davis, C., Rittenhouse, E., and Fisher, L.: Femoropopliteal bypass with externally supported knitted Dacron grafts: A follow-up of 200 grafts for one to twelve years. J. Vasc. Surg., 19:487, 1994.
12. Harker, L. A., and Hanson, S. R.: Graft thrombus formation, detection, and resolution. In Stanley, J. (Ed.): Biologic and Synthetic Vascular Prostheses. New York, Grune & Stratton, 1983, p. 101.
13. Harris, E. J., Jr., Taylor, L. M., Jr., McConnell, D. B., Moneta, G. L., Yeager, R. A., and Porter, J. M.: Improved modern results of axillobifemoral bypass using externally supported PTFE. J. Vasc. Surg., 12:416, 1990.
14. Harris, R. W., Audros, G., Dulawa, L. B., Oblath, R. W., et al.: Successful long term limb salvage using cephalic vein bypass grafts. Ann. Surg., 200:785, 1984.
15. Herring, M., Smith, J., Dalsing, M., Glover, J., Compton, R., Etchberger, K., and Zollinger, T.: Endothelial seeding of polytetrafluoroethylene femoral popliteal bypasses: The failure of low density seeding to improve patency. J. Vasc. Surg., 20:650, 1994.
16. Kalman, P. G., Hosang, M., Johnston, K. W., and Walker, D. M.: The current role for femorofemoral bypass. J. Vasc. Surg., 6:71, 1987.
17. Kenney, D. A., Berger, K., Walker, M. W., and Sauvage, L.: Experimental comparison of the thrombogenicity of fibrin and PTFE flow surfaces. Ann. Surg., 191:355, 1980.
18. Leather, R. P., Shah, D. M., Chang, B. B., and Kaufman, J. L.: Resurrection of the in situ saphenous vein bypass: 1000 cases later. Ann. Surg., 208:435, 1988.
19. Mannick, J. A., Jackson, B. T., Coffman, J. D., and Hume, D. M.: Success of bypass vein grafts in patients with isolated popliteal artery segments. Surgery, 61:17, 1967.
20. Ochsner, J. L., DeCamp, P. T., and Leonard, G. L.: Experience with fresh venous allografts as an arterial substitute. Ann. Surg., 173:933, 1971.
21. O'Donnell, T. F., Farber, F. P., Richmond, D. M., Deterling, R. A., and Callow, A.: Above knee polytetrafluoroethylene bypass graft: Is it a reasonable alternative to the below-knee reversed autogenous vein grafts. Surgery, 94:26, 1983.
22. Okuhn, S. P., Connelly, D. P., Calakos, N., Ferrell, L., Man-Xiang, P., and Goldstone, J.: Does compliance mismatch alone cause neointimal hyperplasia? J. Vasc. Surg., 9:35, 1989.
23. Parodi, J. C., Palmaz, J. C., and Barone, H. D.: Transfemoral intraluminal graft implantation for abdominal aortic aneurysms. Ann. Vasc. Surg., 5:491, 1991.
24. Piotrowski, J. J., Pearce, W. H., Jones, D. N., Whitehill, T., Bell, R., Patt, A., and Rutherford, R. B.: Aorta bifemoral bypass: The operation of choice for unilateral iliac occlusion? J. Vasc. Surg., 8:211, 1988.
25. Reigel, M. M., Hollier, L. H., Pairolero, P. C., and Hallett, J. W., Jr.: Early experience with a new collagen-impregnated aortic graft. Am. Surg., 54:134, 1988.
26. Rutherford, R. R., Flanigan, D. P., Gupta, S. K., Johnston, K. W., Karmody, A., Whittemore, A. D., Baker, J. D., and Ernst, C. B.: Suggested standards for reports dealing with lower extremity ischemia. J. Vasc. Surg., 4:80, 1986.
27. Schmitz-Rixen, T., Megerman, J., Colvin, R. B., Williams, A. M., and Abbott, W. M.: Immunosuppressive treatment of aortic allografts. J. Vasc. Surg., 7:82, 1988.
28. Schulman, M. L., Badhey, M. R., and Yatco, R.: Superficial femoral popliteal veins and reversed saphenous veins as primary femoropopliteal bypass grafts: A randomized comparative study. J. Vasc. Surg., 6:1, 1987.
29. Scott, S. M., Barth, M. G., Gaddy, L. R., and Ahl, E. T., Jr.: The role of circulating cells in the healing of vascular protheses. J. Vasc. Surg., 19:585, 1994.
30. Shue, W. B., Worosilo, M. S., Donetz, A. P., Trooskin, S. Z., Harvey, R. A., and Greco, R. S.: Prevention of vascular prosthetic infection with an antibiotic bonded Dacron graft. J. Vasc. Surg., 8:600, 1988.
31. Sottiurai, V. S., and Arson, R. C.: Ultrastructural studies of arterial grafts. In Bergan, J. J., and Yao, J. S. T. (Eds.): Evaluation and Treatment of Upper and Lower Extremity Circulatory Disorders. New York, Grune & Stratton, 1984, p. 371.
32. Spray, T. L., and Roberts, W. C.: Fundamentals of clinical cardiology: Changes in saphenous veins used as aortocoronary bypass grafts. Am. Heart. J., 94:500, 1977.
33. Stanley, J. C., Ernst, C. B., and Fry, W. J.: Fate of 100 aortorenal vein grafts; characteristics of late graft expansion, aneurysmal dilation, and stenosis. Surgery, 74:931, 1973.
34. Stoney, R. J., DeLuccia, N., Ehrenfeld, W. K., and Wylie, E. J.: Aortorenal arterial autografts. Arch. Surg., 116:1416, 1981.
35. Szilagyi, D. E., Elliott, J. P., Hageman, J. G., Smith, R. F., and Dall'Olmo, C. A.: Biologic fate of autogenous vein implants and arterial substitutes. Ann. Surg., 178:232, 1973.
36. Szilagyi, D. E., Hageman, J. G., Smith, R. F., et al.: Autogenous vein grafting in femoropopliteal atherosclerosis: The limit of its effectiveness. Surgery, 86:836, 1979.
37. Taylor, L. M., Jr., Edwards, J. M., Phinney, E. S., and Porter, J. M.: Reversed vein bypass to infrapopliteal arteries. Ann. Surg., 205:90, 1987.
38. Taylor, L. M., Jr., Edwards, J. M., and Porter, J. M.: Present status of reversed vein bypass: Long term results of a modern series. J. Vasc. Surg., 11:193, 1990.
39. Taylor, L. M., Jr., Park, T. C., Edwards, J. M., Yeager, R. A., McConnell, D. B., Moneta, G. L., and Porter, J. M.: Acute disruption of polytetrafluoroethylene grafts adjacent to axillary anastomoses: A complication of axillofemoral grafting. J. Vasc. Surg., 20:520, 1994.
40. Tollefson, D. F., Bandyk, D. F., Kaebnick, H. W., et al.: Surface biofilm disruption: Enhanced recovery of microorganisms from vascular prostheses. Arch. Surg., 122:38, 1986.
41. Williams, G. M., ter Haar, A., Krajewski, D., Parks, L. C., and Roth, J.: Rejection and repair of endothelium in major vessel transplants. Surgery, 78:694, 1975.
42. Wilson, J. M., Birinyi, L. K., Salmon, R. N., Libby, P., Callow, A. D., and Mulligan, R. C.: Implantation of vascular grafts lined with genetically modified endothelial cells. Science, 244:1344, 1989.
43. Yeager, R. A., Moneta, G. L., Taylor, L. M., Jr., Harris, E. J., Jr., McConnell, D. B., and Porter, J. M.: Improving survival and limb salvage in patients with aortic graft infection. Am. J. Surg., 159:466, 1990.
44. Zilla, P., Deutsch, M., Meinhart, J., Puschmann, R., Eberl, T., Minar, E., Dudczak, R., Lugmaier, H., Schmidt, P., Noszian, I., and Fischleiu, T.: Clinical in vitro endothelialization of femoropopliteal bypass grafts: An actuarial follow-up over three years. J. Vasc. Surg., 19:540, 1994.
The text and refencences above were taken from:
Gregory L. Moneta, John M. Porter. Arterial Substitutes. In: David C. Sabiston, H. Kim Lyerly, ed. Textbook of surgery: the biological basis of modern surgical practice, 15th ed. Philadelphia: W.B. Saunders Company; 1997: 1626-1637.
RECENT STUDIES
Grabenwoeger, M., Fitzal, F., Sider, et al, have investigated the feasibility of transmural capillary ingrowth into the inner surface of biosynthetic vascular prostheses (Omni-flow, BioNova, Melbourne, Australia) through perforations created by an excimer laser, thus inducing an endothelial cell coverage and concluded that spontaneous endothelialization of biosynthetic vascular prostheses can be achieved by transmural capillary ingrowth through perforations in the wall of the prostheses in an experimental sheep model. 1
Blumenberg, R M, Anderson, J M, et al, have evaluated the effect of heparin coating on patency rate and intimal hyperplasia in small synthetic vascular grafts and concluded that heparin coating seems to be beneficial for graft patency. 2
Yoneyama T, Ito M, et al, have prepared a small diameter vascular prosthesis functioning without pseudointima formation and suggested that the SPU(segmented polyurethane) /MPC(2-methacryloyloxyethyl phosphorylcholine) 10.0 wt% prosthesis, functioning without a pseudointima, possesses a stronger nonthrombogenicity and would be more applicable for clinical use. 3
Blumenberg, R M, Anderson, J M, et al, have histologically evaluated Dacron® and PTFE graft material explanted from humans After 4 to 20 Years in vivo. Seven knitted Dacron(R) and four PTFE prosthetic arterial grafts, when examined histopathologically following explantation 41/2 to 20 years following placement in 10 patients, have demonstrated no evidence of mechanical or chemical biodegradation. 4
Hotoveli-Salomon et al, have evaluated the short-term performance of selectively biodegradable filament-wound vascular prostheses, comprising elastomeric poly(ether urethane) (Lycra) scaffolds and flexible, hydrophilic biodegradable coatings and concluded that the improved mechanical properties and enhanced healing of the highly porous filament-wound Lycra scaffold graft coated with hydrophilic biodegradable PELA has the potential of being a highly effective small caliber prosthetic graft. 5
S Post, T Kraus, et al, have compared the patency of PTFE (Polytetrafluoroethylene) and unsealed knitted Dacron femoro-popliteal bypasses and found that PTFE and Dacron are equally suitable for femoro-popliteal bypass. 6
M Prager, P Polterauer, et al, have compared knitted gelatin-coated Dacron bifurcation grafts, knitted collagen-coated Dacron grafts, and stretch polytetrafluoroethylene (PTFE) grafts and found no difference between the 3 graft materials in long-term patency. When both Dacron grafts were compared collectively with stretch PTFE, a difference has been found in infection rate: Dacron 3% (6/209) versus PTFE 0% (0/106); P <.03. 7
R. Zippel, L. Wilhelm, et al, have investigated the specific humoral immune response to three different polyester (Dacron) prostheses in pigs and concluded that specific IgG antibodies, reflecting the inflammatory response, might be not only a parameter for testing biomaterials but also for determining individual bio(in)compatibility for long-term biomaterial function. 8
Karube, N., Takanashi, et al, have studied clinical long-term results of vascular prosthesis sealed with Fragmented Autologous Adipose Tissue(FAT). From the results of this clinical trial, they have concluded that the FAT grafts were acceptable as vascular prostheses for ischemic lower extremities. 9
To test the hypothesis that the expression of leukocyte adhesion molecules by the endothelial cells is modulated by the underlying polymer surface, and to correlate these findings with several polymer surface properties, polymer surface modification was suggested. The surface properties of available biomaterials (like ePTFE or Dacron) can be modified by Radio Frequency Glow Discharge treatment.
Several polymer surfaces have been created from commonly used biomaterials, like PET and Teflon, using gas plasma techniques. Endothelial cell growth on these plasma-treated surfaces has been improved, as compared to the non-treated surfaces, and these surfaces are therefore under further investigation. Liquid flow, and its ‘shear stress’, has a strong positive effect on endothelial-layer development compared to growth under static conditions. It prepares the endothelium for blood-flow conditions and, among others, results in an improved binding of the endothelial cells to the polymer surface. 10
A Cornell University fiber/biomedical material scientist has made two breakthroughs, both just recently approved by the U.S. Patent Office, to help the health care industry and medical patients: a non-toxic method for sterilizing biodegradable medical materials and devices, and an innovative chemical process that could make implants "biologically active" to help promote healing and fight off disease. The first invention involves modifying the currently used gamma irradiation sterilization process with extremely low temperatures (-192 degrees C) and a very strong vacuum. This method could replace the current ethylene oxide gas method of sterilizing biodegradable materials and devices, which is tedious, time-consuming and toxic to workers. The second invention a new chemical process to attach nitric oxide and its derivatives onto biomaterials. Such bonding could transform implants from having "passive" roles to becoming "biologically active" in helping human body repair and disease resistance.One area of interest in our laboratories at The University of Texas is the vascular system. Researchers have fabricated a model of a blood vessel for the purpose of understanding vascular development and regulation and for the designing of tissues suitable for use as vascular grafts in patients. Blood vessel substitutes, as needed in heart bypass surgery for example, are in high demand as more than 500,000 coronary artery bypass operations are performed annually in the United States alone. The best results are obtained if one of the patient's own vessels is grafted, but if an autologous vessel cannot be used, a prosthetic vessel may be implanted. Vein grafts require that a vein be removed from another part of the patient (usually the leg), requiring multiple surgical procedures. It is known that 20 percent to 30 percent of all patients do not have usable veins for this procedure, usually due to previous vein harvest, amputation, or other medical conditions. In addition, as synthetic grafts are foreign to the body, they may pose long-term health risks. Although synthetic materials can be successfully used in the formation of large-diameter vessels, they are not as useful for the construction of the small-diameter grafts required for many heart bypass surgeries, as the grafts are more prone to clotting. Many researchers believe that modification of existing grafts with a lining mimicking the natural endothelial cell lining found on blood vessels in the body would minimize clotting by providing a barrier between blood-clotting components and the graft material. The use of endothelial cells derived from the patient would be ideal since an immune response would be avoided. However, harvesting endothelial cells from a patient for use on a vascular graft is clinically infeasible in emergencies. To overcome this limitation, researhers are studying the use of animal endothelial cells which are readily available but need to be modified to suppress the immune reaction. Animal endothelial cells are recognized as foreign and then rapidly rejected by the body's immune system primarily due to a single sugar group expressed on the surface of the animal cells, which is recognized by natural antibodies in humans. Genetic modification of the pathway leading to expression of this sugar group in other cells has previously been shown to successfully reduce the sugar group's surface display and to suppress the immune response. By exploiting a similar molecular strategy, researhers are taking steps toward the development of a universal endothelium that would not be rejected by the body and could be combined with a suitable scaffold to prevent clotting of a vascular graft. 11
COMMERCIAL PRODUCTS
TRICOGEL® grafts are composed of knitted polyester double velour vascular prostheses and Porcine gelatin. The two elements are bonded by surface adhesion.TRICOGEL® grafts are characterized by:light yellow color ; intraoperative tightness – no preclotting ; hydrophilia which can be observed as an instant moistening of the surface with patient’s blood and as sweating. The blood stream does not dissolve nor washes away the gelatin but causes the gelatin film to swell, which makes a better tightness. antithrombogenicity – a coat of networked gelatin is hemocompatibile until its complete biodegradation and does not activate any platelets, rigidity when dry, which fades after contact with liquids (blood) ; high mechanical strength ; elasticity and prevention to stretching and bending the gelatin film which take place when the graft is taken through the canals in the abdominal cavity; low mass ; no fraying and easy handling ; clear marking line. TRICOGEL® grafts are prestretched and sold as a sterile biomaterial. Grafts are sold as sterile. Colored marker on the package inside the box indicates radiation sterilization performed on the material. The marker darkens under radiation. Grafts cannot be resterilized.Sterility period guaranteed by the manufacturer is 3 years. It is recommended to suture TRICOGEL® grafts with synthetic, non-resorbant, polyester or polypropylene sutures using atraumatic semicircular needles. Sutures 3-0 are recommended for suturing and aortic anastomoses and brachiocephalic trunk, whereas sutures 4-0 ought to be used for suturing iliac arteries. 12
THORACIC, ABDOMINAL, AORTO ILLIAC - ANEURYSMAL AND OCCLUSIVE DISEASE
Hemashield GoldTM Textile Vascular Grafts, Improving On The Gold Standard, exceptional handling characteristics, collagen, the natural healing interface , impregnation, for a controlled healing response, specially formulated for the healing advantage. 13
VanguardTM Endovascular Aortic Grafts, Opening Up New Possibilities in Less Invasive AAA Treatment, less invasive treatment modality , fully supported and modular for procedural flexibility, proprietary, strong, thin wall, seamless graft material, self-expanding stent scaffold with unique flexibility. 13
PassagerTM Endovascular Peripheral Grafts,A less invasive treatment option, designed to prevent neo-intimal hyperplasia, proprietary endovascular graft material, stent scaffold with unique flexibility. 13
PERIPHERAL VASCULAR DISEASE AND VASCULAR ACCESS
Meadox ExxcelTM ePTFE Vascular Grafts, providing superior balance of handling, strength, and healing through: Proprietary cross-helical yarn wrap, macroscopic surface texture, consistent microstructure, available in diverse sizes and configurations. 13
REFERENCES
Grabenwoeger, M., Fitzal, F., Sider, J., Csekoe, C., Bergmeister, H., Schima, H., Husinsky, W., Boeck, P., Wolner, E., Grabenwöger M, Fitzal F, Sider J, Csekö C, Bergmeister H, Schima H, Husinsky W, Böck P, Wolner E : Endothelialization of Biosynthetic Vascular Prostheses After Laser Perforation : ANNALS OF THORACIC SURGERY, SUPP/1(6):S110-S114 1998.
2.Blumenberg, R. M., Anderson, J. M., Gelfand, M. L., Skudder, P. S., Bowers, C. A.: CIRCULATION (AMERICAN HEART ASSOCIATION) 98(19 Suppl):II319-23; 1998.
3.Yoneyama, T., Ito, M., Ken-ichi, S., Ishihara, K., Nakabayashi, N.: Small Diameter Vascular Prosthesis with a Nonthrombogenic Phospholipid Polymer Surface: Preliminary Study of a New Concept for Functioning in the Absence of Pseudo- or Neointima Formation: ARTIFICIAL ORGANS,24(1):23-28 2000.
4.Blumenberg, R. M., Anderson, J. M., Gelfand, M. L., Skudder, P. S., Bowers, C. A. : Histologic Evaluation of Dacron® and PTFE Graft Material Explanted from Humans After 4 to 20 Years In Vivo : VASCULAR SURGERY , 34(6):505-512 2000.
5.Hotoveli-Salomon, Anna, Schwalb, Herzl, Hellener, Gunnar R., Marom, Gad, Stern, Theodor: Novel synthetic selectively degradable vascular prostheses: A preliminary implantation study: JOURNAL OF SURGICAL RESEARCH ,95(2):152-160 2001.
6.S. Post, T. Kraus, U. MÜllerReinartz, C. Weiss, H. Kortmann, A. Quentmeier, M. Winkler: Dacron vs Polytetrafluoroethylene Grafts for Femoropopliteal Bypass:a Prospective Randomised Multicentre Trial:: EUROPEAN JOURNAL OF VASCULAR AND ENDOVASCULAR SURGERY , 22(3):226-231 2001.
7.M Prager, P Polterauer, HJ Bahmig, O Wagner et al.: Collagen versus gelatin-coated Dacron versus stretch polytetrafluoroethylene in abdominal aortic bifurcation graft surgery: Results of a seven-year prospective, randomized multicenter trial: SURGERY -OXFORD- 130(3):408-414 2001
8.R. Zippel, L. Wilhelm, A. Koch, F. Marusch, M. Schlosser. Antigenicity of Polyester (Dacron) Vascular Prostheses in an Animal Model. EUROPEAN JOURNAL OF VASCULAR AND ENDOVASCULAR SURGERY , 21(3):202-207 2001.
9.Karube, N., Takanashi, Y., Noishiki, Y., Yamazaki, I., Kosuge, T., Soma, T., Ichikawa, Y. The Clinical Long-Term Results of Vascular Prosthesis Sealed with Fragmented Autologous Adipose Tissue. ARTIFICIAL ORGANS , 25(3):218-222 2001.
Figure 1.Close-up photograph (×50) of a woven Dacron prosthetic vascular graft. This graft is relatively impervious to blood because of the tightness of the weave.
Figure 2.Close-up photograph (×50) of a knitted Dacron prosthetic vascular graft. The large openings between the knitted yarns make this graft relatively permeable to blood, and the graft must be preclotted before use in order to fill the interstices with fibrin.
Figure 3.Photograph (×50) of the external surface of a woven double velour Dacron graft. The striking difference in the surface texture compared with that of a standard knitted or woven Dacron prosthesis is obvious. The velour configuration promotes rapid fibrous anchoring of the graft to surrounding tissues.
Figure 4. Photograph of a standard knitted Dacron bifurcation graft. The crimp pattern is clearly seen. The stripes aid the surgeon in proper graft positioning and diminish the likelihood of unrecognized axial rotation of the graft limb when placed in the aortofemoral position
Figure 5. Angiogram of an aortofemoral knitted bifurcation Dacron graft. The graft extends from the infrarenal aorta to the common femoral arteries.
Figure 6.Photomicrograph (×1000) of a PTFE graft. The dark areas are PTFE nodes interconnected by many PTFE fibrils. Average internodal distance in grafts in clinical use is now 40µ
Figure 7. A defective knitted Dacron prosthesis resulted in this false aneurysm (arrow) of the midportion of a femorofemoral graft.
Figure 8. A Dacron graft is mounted on an expandable metallic stent.
Hemashield GoldTM Textile Vascular Grafts
VanguardTM Endovascular Aortic Grafts
PassagerTM Endovascular Peripheral Grafts
Meadox ExxcelTM ePTFE Vascular Grafts
TRICOGEL®
TRICOGEL®
DALLON® H
This site is best viewed with a resolution of 800x600. Contact webmaster for site related problems.
