Hemodialysis Vascular Access Dysfunction: A Cellular and Molecular Viewpoint
Prabir Roy-Chaudhury*,
Vikas P. Sukhatme and
Alfred K. Cheung
* University of Cincinnati Medical Center, Cincinnati, Ohio; Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts; and Veterans Affairs Salt Lake City Healthcare System and the University of Utah, Salt Lake City, Utah
Address correspondence to: Dr. Prabir Roy-Chaudhury, Division of Nephrology, MSB G-251, University of Cincinnati, 231 Albert Sabin Way, Cincinnati, OH 45267-0585. Phone: 513-558-4006; Fax: 513-558-4309; E-mail: prabir.roychaudhury{at}uc.edu
Hemodialysis vascular access dysfunction is a major cause ofmorbidity and hospitalization in the hemodialysis population.The major cause of hemodialysis vascular access dysfunctionis venous stenosis as a result of neointimal hyperplasia. Despitethe magnitude of the clinical problem, however, there has beena paucity of novel therapeutic interventions in this field.This is in marked contrast to a recent plethora of targetedinterventions for the treatment of arterial neointimal hyperplasiaafter coronary angioplasty. The reasons for this are two-fold.First there has been a relative lack of cellular and molecularresearch that focuses on venous neointimal hyperplasia in thespecific setting of hemodialysis vascular access. Second, therehave been inadequate efforts by the nephrology community totranslate the recent advances in molecular and interventionalcardiology into therapies for hemodialysis vascular access.This review therefore (1) briefly examines the different formsof hemodialysis vascular access that are available, (2) describesthe pathology and pathogenesis of hemodialysis vascular accessdysfunction in both polytetrafluoroethylene grafts and nativearteriovenous fistulae, (3) reviews recent concepts about thepathogenesis of vascular stenosis that could potentially beapplied in the setting of hemodialysis vascular access dysfunction,(4) summarizes novel experimental and clinical therapies thatcould potentially be used in the setting of hemodialysis vascularaccess dysfunction, and, finally, (5) offers some broad guidelinesfor future innovative translational and clinical research inthis area that hopefully will reduce the huge clinical morbidityand economic costs that are associated with this condition.
Introduction and Types of Hemodialysis Vascular Access
Performance of a successful hemodialysis procedure requiresa functional vascular access. Unfortunately, there have beenno major advances in the field of hemodialysis vascular accessfor the past three decades, which probably has contributed tohemodialysis vascular access dysfunction being one of the mostimportant causes of morbidity in the hemodialysis population(1). With the use of Medicare data, it has been estimated thatvascular access dysfunction is responsible for 20% of all hospitalizationsin the hemodialysis population (2). In 2001, vascular accesscomposed 7.5% of the $14 billion spent by Medicare on the ESRDprogram (approximately $1 billion per annum) (3).
There currently are three main forms of hemodialysis vascularaccess: (1) The native arteriovenous fistula (AVF), (2) thepolytetrafluoroethylene (PTFE) graft, and (3) the cuffed double-lumensilicone catheter. Each of these forms of hemodialysis vascularaccess has its own specific problems.
Native AVF: Once mature and functional, AVF are the preferredform of hemodialysis vascular access because of their relativelack of infection and thrombosis (Figure 1) (4). They are usuallycreated through the surgical anastomosis of the radial (wrist)or brachial (elbow and upper arm) artery to the cephalic vein.Their two major complications are an initial failure to mature(primary nonfunction) and a later venous stenosis followed bythrombosis. A pooled data analysis of AVF survival that wasperformed at the time of the Dialysis Outcomes Quality Initiative(4) suggested a primary patency of 85% for AVF at 1 yr and 75%at 2 yr (Figure 1). These data, however, exclude fistulae thatdid not mature adequately to support hemodialysis. Primary nonfunctionor rates of failure to mature up to 50% have been reported bysome centers, particularly when an aggressive fistula placementpolicy is enforced (5). Therefore, primary patency rates aslow as 43% (6) and 56% (7) have been reported for AVF in someinstances. In addition, Miller et al. (5) showed a higher primarynonfunction rate in forearm fistulae as compared with upperarm fistulae (59 versus 34%). This difference was greatest inwomen, patients with diabetes, and patients who were older than65 yr, suggesting that these patient groups perhaps should haveupper arm fistulae placed as the primary access procedure. Finally,it is important to point out that the United States has an extremelylow AVF rate (Figure 2) as compared with other industrializedcountries. Aggressive clinical efforts are under way to improvethis shortcoming (8).
PTFE grafts: Arteriovenous PTFE dialysisgrafts remain the mostcommon form of hemodialysis vascularaccess in the United States.Although relatively easy to placeand ready to use, they haveextremely high rates of stenosis,thrombosis, and infection.The pooled data for the DialysisOutcomes Quality Initiativeanalysis suggested a primary patencyrate of 50% at 1 yr (Figure 1).Other authors have reportedprimary patency rates as low as23% at 1 yr and 4% at 2 yr (9).
Cuffed double-lumen silicone catheters: These should be avoidedat all costs because of an extremely high incidence of thrombosisand infection, except as a temporary measure or when the lifeexpectancy of the patient is short. Unfortunately, this oftenis difficult to accomplish at a clinical level because thisis the most convenient way to obtain immediate dialysis access.
Figure 1. Comparison of patency of arteriovenous fistula (AVF) versus polytetrafluoroethylene (PTFE) dialysis grafts. (a) Unassisted primary patency of native AVF versus PTFE grafts. (b) Cumulative patency in the setting of an active monitoring and intervention program. Note that with prospective monitoring, the cumulative patency of PTFE grafts is similar to that for native AVF but at the cost of a six-fold increase in the intervention rate. Graphs are derived from the summed data analysis for the Dialysis Outcomes Quality Initiative panel. Modified from reference (4). The AVF survival curves exclude primary nonfunction (failure to mature) and show the failure rate only for AVF that matured enough to be used for hemodialysis. Currently, many centers follow an aggressive policy of early fistula salvage, and between 10 to 50% of all fistulae that are placed may need some intervention, depending on the aggressiveness of fistula placement.
Figure 2. Dialysis Outcomes Practice Patterns Study (DOPPS) data. Note the dismal native AVF prevalence rate and the high PTFE graft prevalence rate in the United States as compared with Europe and Japan. (DOPPS data as of September 2003; courtesy of Dr. Rajiv Saran, University of Michigan, Ann Arbor, MI.)
This brief comparison of the three major forms of hemodialysisvascular access leaves us with the following messages. (1) Weneed to refine and popularize the surgical techniques and maximizethe clinical use of native fistulae by optimizing logisticaland practice pattern issues while developing new therapies topromote fistula maturation and reduce venous stenosis. (2) Thestenosis and thrombosis rate associated with PTFE dialysis graftsshould be reduced so that they can become truly a long-lastingform of dialysis access. (3) More effective anti-infective andantithrombotic therapies for cuffed double-lumen silicone cathetersshould be developed so that immediate blood access can be obtainedif necessary, without significant morbidity.
This review focuses on the failure of the native or PTFE arteriovenousaccess. It critically examines the current state of knowledgewith regard to the pathology and pathogenesis of their dysfunction;identifies novel concepts and therapies that could be appliedto this field; and suggests strategies by which recent advancesin molecular biology, biomaterials, and drug delivery couldbe applied to hemodialysis vascular access dysfunction.
Pathology and Pathogenesis of Native Fistula and PTFE Graft Failure
Pathology of Native Fistula and PTFE Graft Failure
The two main causes of native AVF failure are an initial failureto mature followed by a later venous stenosis. The exact pathologythat results in primary nonfunction of these fistulae is unclear,but the characteristic lesion is a juxta-anastomotic stenosis(Figure 3a). It is unclear, however, whether the primary factorthat causes this stenosis is venous constriction or venous neointimalhyperplasia. The pathology of late venous stenosis in nativefistulae is believed to be similar to that of venous stenosisin the setting of PTFE dialysis grafts (see below). The stenosesoccur at or around the anastomotic region in wrist fistulae(Figure 3b) and in the proximal vein in the setting of fistulaethat are constructed at the elbow (Figure 3c) (10). Recent workalso documents a high incidence of inflow arterial stenosis(11). Stenosed AVF have been shown to have venous neointimalhyperplasia composed of smooth muscle cells with expressionof cytokines and mediators such as endothelin, PDGF, and TGF-(the last has been shown to co-localize with markers of oxidativestress) within the media and intima (12, 13).
Figure 3. AVF problems. (a) The classic angiographic picture of a juxta-anastomotic stenosis (arrows) resulting in a failure to mature (courtesy of Dr. Tony Samaha, Cincinnati, OH). These lesions can be diagnosed and treated aggressively (angioplasty), with excellent results. (b and c) The sites of venous stenosis for native AVF at the wrist (b) and at the elbow (c). Adapted from reference (10) with the addition of recent data (7), on the increasing incidence (recognition) of arterial stenoses.
The major cause of PTFE graft failure is a venous stenosis thatresults in thrombosis, either at the graft–vein anastomosisor in downstream or proximal vein (14). More recently, thereseems to be a rise in stenoses at the graft–artery anastomosisand within the intragraft region as well (11) (Figure 4). Venousstenosis in PTFE dialysis access grafts is due to venous neointimalhyperplasia, which is characterized by the presence of smoothmuscle cells, myofibroblasts, and microvessels within the venousneointima. In addition, there is a significant amount of adventitialangiogenesis and a large number of macrophages that line theperigraft region (a major difference from the venous stenosisthat occurs in native fistulae) (15–17). Our group alsohas demonstrated the presence of cytokines, such as PDGF, vascularendothelial growth factor, and basic fibroblast growth factor,and of matrix proteins, such as collagen and tenascin, withinthe lesion of venous neointimal hyperplasia (15) (Figure 5).
Figure 4. Sites of venous stenoses in PTFE dialysis grafts. (a) The sites of venous stenosis in PTFE dialysis grafts. Note the preponderance of lesions at the graft–vein anastomosis or within 6 to 10 cm of the anastomosis and also at the arterial anastomosis. (b) Angiogram of a PTFE dialysis graft with a developing pseudoaneurysm (arrowhead) and stenosis (arrow) at the graft–vein anastomosis (courtesy of Dr. Tom Vesely, Malinckrodt Institute of Radiology, St. Louis, MO).
Figure 5. Venous neointimal hyperplasia in PTFE dialysis grafts (human samples). (a) PTFE graft. Note the significant venous neointimal hyperplasia (extent of arrow) between the graft (G) and the lumen (L). (b) Downstream (proximal) vein (N, neointima; M, media), smooth muscle actin (SMA) demonstrates that the majority of cells in downstream vein are smooth muscle cells. (c) Downstream vein (neointima). Note the prominent angiogenesis within the neointima (arrows) as assessed by this endothelial cell marker. (d) Downstream vein (neointima). High-power view of a microvessel within the neointima of downstream vein. Note the distinct co-localization of blue (endothelial) and brown (proliferating) cells indicating active endothelial cell proliferation (angiogenesis). (e) Upstream graft (neointima). High-power view of a macrophage giant cell adjacent to the neointimal surface of PTFE graft (G). Also note the large number of macrophages in this area (thin arrows). (f) Downstream vein (media and neointima). There is strong expression of this cytokine in the venous media (M) and by smooth muscle cells/myofibroblasts within the neointima (N; bar). Adapted from reference (15). Magnifications: x200 in a (hematoxylin and eosin); x400 in b (SMA), c (von Willebrand factor [vWF]), and f (PDGF); x800 in d (vWF and Ki67); x2000 in e (PG-M-1).
Pathogenesis of AVF and PTFE Graft Failure
The pathogenesis of early native arteriovenous fistula failure(juxta-anastomotic stenosis) is complex and multifactorial (Figure 6).Causative factors include a small artery (<1.5 to 2 mm) anda small vein (<2.0 to 2.5 mm), surgical manipulation andless-than-ideal technique, previous venipunctures, the developmentof accessory veins that direct blood away from the primary venousdrainage channel, hemodynamic stressors (see below), and a possiblegenetic predisposition to vasoconstriction and neointimal hyperplasiaafter endothelial and smooth muscle injury (18, 19). It stillis unclear whether vascular constriction, neointimal hyperplasia,or a combination of the two factors is responsible for the earlymaturation failure of native fistulae.
Figure 6. From pathogenesis to pathology to novel therapies. This figure identifies the different pathogenetic mechanisms that result in dialysis access stenosis and directs attention to potential novel therapies. The pathogenetic factors include hemodynamic and surgical stressors, inflammatory stimuli from dialysis needles and PTFE graft material, and the unavoidable vascular injury that occurs at the time of angioplasty (see text for a more detailed discussion). Novel therapeutic modalities include perivascular drug delivery, drug-eluting stents, coated grafts, novel balloons (picture of Conquest balloon courtesy of Bard Peripheral Vascular), and better final surgical anatomy (see text for a more detailed discussion).
The pathogenesis of venous neointimal hyperplasia in PTFE dialysisgrafts and late failure in native fistulae is better understoodand comprises a cascade of events that are best discussed underthe heading of upstream and downstream events. Upstream eventsprimarily are the factors that are responsible for endothelialand smooth muscle injury, which then set into motion the complexinterplay of cells, cytokines, and mediators (downstream events)that result in venous neointimal hyperplasia.
The upstream events in the pathogenesis of venous neointimalhyperplasia in the setting of dialysis grafts and fistulae include(1) hemodynamic stress at the graft–vein or artery–veinanastomosis as a result of a combination of low shear stress,turbulence, and compliance mismatch between noncompliant graft/arteryand compliant vein (20, 21); (2) surgical injury at the timeof creation of the arteriovenous conduit (this perhaps couldbe most significant in the setting of AVF, in which the veinoften is stretched and manipulated (Dr. Klaus Konner, Universityof Cologne, Germany, personal communication, March 18, 2004);(3) the presence of the PTFE graft itself, which has been shownto attract in macrophages, producing a plethora of cytokines(15); (4) graft injury from dialysis needles (this probablyis most relevant in the setting of intragraft stenoses); and(5) the presence of uremia, which has been shown to exacerbateendothelial dysfunction (22) and predispose to venous neointimalhyperplasia even before the creation of an arteriovenous dialysisaccess. After development of an initial stenosis, it is likelythat the most important factor that predisposes to repeat stenosesis endothelial and smooth muscle cell injury at the time ofangioplasty for the treatment of the initial stenosis. Changet al. (23) demonstrated a significant increase in the proliferationindex within the venous neointima and media in patients withaggressive restenotic lesions as compared with patients withprimary stenotic lesions. Their study emphasizes the need forconcomitant administration of antiproliferative therapy at thetime of balloon angioplasty of dialysis access grafts and fistulae(as is currently the standard of care after coronary angioplasty).
The downstream events are essentially a response to endothelialand smooth muscle cell injury secondary to the upstream events.According to the "traditional theory" of neointimal hyperplasia,endothelial and smooth muscle injury results in the migrationof smooth muscle cells and myofibroblasts from the media intothe intima, where they proliferate and form the lesion of venousneointimal hyperplasia. This process of injury followed by migrationand proliferation is orchestrated by a large number of mediators,which include the cell-cycle regulators (p27 and p16, retinoblastomaprotein, p38 mitogen-activated protein kinase); cytokines (PDGF,basic fibroblast growth factor, and TNF-); chemokines (monocytechemoattractant protein-1 and RANTES); vasoactive molecules(nitric oxide [NO] and endothelin); adhesion molecules (intercellularadhesion molecule-1 and P-selectin); and molecules such as osteopontin,apolipoprotein E, matrix metalloproteinase-2, and human hepatocytegrowth factor (24). It should be emphasized, however, that mostof this information is derived from arterial angioplasty modelsrather than from vascular (venous) anastomosis models.
At a clinical level, dialysis access grafts and fistulae developstenosis and thrombosis far more aggressively as compared witharterial grafts or even interposition saphenous vein conduitsin coronary artery bypass surgery. This may be due to the followingfactors. (1) At an anatomic level, veins tend to have a lesswell-defined internal elastic lamina, which could predisposeto the migration of smooth muscle cells and myofibroblasts fromthe media to the intima. (2) At a physiologic and molecularlevels, veins tend to produce less NO and prostacyclin, whichcould predispose to endothelial injury (25). Recent studieshave also demonstrated significant differences in gene expressionbetween normal arteries and veins (26). Whether differencesin the expression of these genes translates into a more aggressiveresponse to injury is unknown. (3) Puncture of dialysis graftsand fistulae could cause platelet thrombi and cytokine release(27). (4) The presence of uremia could predispose to endothelialdysfunction and stenosis.
It also is important to identify differences in the responseof arteries and veins to vascular injury. This could allow usto develop novel therapies that effectively target venous stenosisand also identify currently available therapies that are mostlikely to be effective in the context of venous stenosis. Forexample, recent work by our groups has demonstrated that venoussmooth muscle cells are more sensitive to the effects of antiproliferativeagents (28) but less sensitive to the effects of radiation ascompared with arterial smooth muscle cells (29).
Finally, we need to mention that in the past few years, a numberof excellent animal models of arteriovenous graft and nativefistula stenosis that closely mimic the clinical lesion havebeen developed. This has greatly helped our understanding ofthe pathology and pathogenesis of arteriovenous stenosis andclearly will be a valuable resource as we move forward to testout novel therapies in this field (30–33).
Novel Concepts in the Pathology and Pathogenesis of Vascular Injury and Stenosis: Relevance to Hemodialysis Vascular Access Dysfunction
Importance of Vascular Remodeling
The functional determinant of hemodialysis vascular access dysfunctionis luminal cross-sectional area at the site of stenosis (Figure 7).This is dependent not only on the magnitude of neointimal hyperplasiabut also on the pattern of vascular remodeling (vasoconstrictionor vasodilation) (34). Indeed, studies in coronary angioplastymodels have suggested that adverse vascular remodeling or vasoconstrictionis responsible for >50% of the final luminal stenosis (34).The factors that are responsible for adverse (rather than beneficial)vascular remodeling are unknown, but adventitial angiogenesisand scar formation (adventitial fibroblasts) are thought toplay a role (35, 36). Therefore, the ideal therapy for vascularstenosis is likely to be an intervention that can block bothadverse vascular constriction (adverse remodeling) and neointimalhyperplasia. This hypothesis is supported by the recent successof drug-eluting stents for the treatment of coronary stenosis(see below) (37, 38).
Figure 7. Adventitial remodeling. The degree of luminal stenosis depends on both the magnitude of neointimal hyperplasia and the degree of vascular remodeling. With the same amount of neointimal hyperplasia, vascular constriction and unfavorable remodeling results in luminal stenosis, whereas favorable remodeling in b prevents the occurrence of luminal stenosis. A similar situation is described in c and d. In both panels, the white area is the lumen. The area in black is the neointima, which is bordered on the outside by the internal elastic lamina and on the inside by the lumen. The hatched area comprises the adventitia and the media. Note that the luminal (white) area in both c and d are identical, despite that d has much less neointima (black area). The reason for this is adverse vascular remodeling in d, which has resulted in a decrease in the area enclosed by the external elastic lamina. This latter parameter is a good indicator of the amount of vascular or adventitial remodeling. Adapted from reference (24).
Alternative Origins for Neointimal Cells (Adventitia and Bone Marrow) Adventitia-Derived Neointimal Cells: A Role for Fibroblasts and Myofibroblasts.
The traditional view on the pathogenesis of neointimal hyperplasiahas emphasized the migration of differentiated contractile smoothmuscle cells from the media to the intima, where they proliferateand contribute to the formation of neointimal hyperplasia. Anumber of groups, however, have shown that after experimentalcoronary angioplasty or saphenous vein grafting, there is amigration of fibroblasts from the adventitia, through the media,and into the intima, where these cells acquire the phenotypeof myofibroblasts (expressing smooth muscle actin) and contributeto the final neointimal volume (Figure 8) (39–41). Ourown work supports the presence of a similar paradigm in thesetting of PTFE dialysis grafts and fistulae in that almost50% of neointimal cells at the graft–vein anastomosisof PTFE dialysis grafts are fibroblasts and myofibroblasts ratherthan contractile smooth muscle cells (42). Although it is possiblethat these cells originally were contractile differentiatedsmooth muscle cells that dedifferentiated, the presence of fibroblastswithin the actual graft material suggests that there is an activemigration of these cells from the adventitial side. The possiblerole of the adventitia, both in vascular remodeling and as asource of neointimal cells, emphasizes that the adventitia isnot an innocent bystander in the pathogenesis of neointimalhyperplasia. Rather, these concepts make a strong case for thedevelopment of therapeutic interventions that (1) focus on theadventitia and (2) target multiple cell types (fibroblasts,myofibroblasts, and smooth muscle cells) instead of only thedifferentiated contractile smooth muscle cell.
Figure 8. Adventitial cells migrate to the intima. High-power view of the stenotic venous limb, which has been used to depict a cartoon representation of the migration of adventitial fibroblasts from the adventitia, through the media, and into the intima, where they can acquire the phenotype of myofibroblasts or smooth muscle cells. Magnification, x800 (vimentin).
Bone Marrow–Derived Neointimal Cells (Figure 9).
Recent data also suggests a role for bone marrow derived smoothmuscle progenitor cells in the pathogenesis of neointimal hyperplasia.Thus Sata et al. (43) have shown that up to 60% of both endothelialand smooth muscle cells within the lesion of neointimal hyperplasiaafter femoral angioplasty are bone marrow-derived cells. Bonemarrow-derived cells that have acquired a smooth muscle phenotypehave also been identified in a mouse model of venous neointimalhyperplasia (44). From a therapeutic standpoint, the local perivascularadministration of sirolimus has been shown to have a beneficialeffect on neointimal hyperplasia by reducing the number of circulatingsmooth muscle progenitor cells within the lesion of neointimalhyperplasia. Unfortunately, sirolimus also reduced the numberof endothelial progenitor cells (see below) which resulted ina decrease in endothelialization (45).
Figure 9. Smooth muscle progenitor cells contribute to neointimal hyperplasia. Smooth muscle progenitor cells have been shown to contribute significantly to total neointimal volume. Adapted from reference (43). Left figure courtesy of Dr. M. Kutryk.
Endothelial Progenitor Cells and Vascular Repair.
Endothelial progenitor cells (EPC) are circulating bone marrow–derivedcells that express both hematopoietic (CD34) and endothelialcell markers (vascular endothelial growth factor receptor 2)(46, 47). In addition to promoting angiogenesis, an importantrole of EPC seems to be the rapid endothelialization of regionsof vascular injury (Figure 10) (48, 49). For example, the infusionof EPC in the setting of angioplasty or surgical graft placementresults in enhanced endothelialization, which translates intoa reduction in neointimal hyperplasia (50, 51). This is in keepingwith previous data that demonstrate an inhibitory effect ofendothelialization and endothelial-derived factors such as NOon smooth muscle cell proliferation and migration (52–54).Most important, a number of agents can enhance the mobilizationof EPC from the bone marrow (statins, erythropoietin, GCSF,and matrix metalloproteinase-9) (48). Indeed, Kong et al. (55)demonstrated that GCSF therapy results in an increase in endothelializationand a reduction in neointimal hyperplasia in a mouse angioplastymodel.
Figure 10. Endothelial progenitor cells (EPC). EPC are produced in the bone marrow and can be mobilized by a number of different factors, including GCSF (a). EPC function as the repair mechanism for the endothelium. (b) EPC bind to injured endothelium possibly to prevent vasospasm, thrombosis, and neointimal hyperplasia. (c) Increased endothelial coverage of the region of vascular injury in mice that were treated with GCSF after balloon angioplasty. This increase in endothelial coverage translated into a reduction in neointimal hyperplasia in mice that were treated with GCSF (d) as compared with untreated mice (e). Adapted from references (48) (a), (49) (b), and (55) (c through e).
The collective data presented in this section suggest a completelynew paradigm for both the pathogenesis and the therapy of neointimalhyperplasia. Specifically, it seems that therapies that (1)prevent the adventitial response to injury and (2) optimizethe balance between smooth muscle progenitor cells (less isgood) and endothelial progenitor cells (more is good) may havethe best chance of inhibiting vascular stenosis as a resultof neointimal hyperplasia. In contrast, traditional therapiesthat aim to prevent the migration and proliferation of smoothmuscle cells from the media to the intima may not be as effectiveat blocking neointimal hyperplasia as a result of the presenceof these alternative pathogenic pathways.
Possible Therapies for Hemodialysis Vascular Access Dysfunction: From Currently Available Intervention to Future Innovation
The information on the pathology and the pathogenesis of vascularstenosis, presented in the previous two sections, helps to identifypotential therapies that could be used to prevent hemodialysisvascular access dysfunction. The focus is on the use of therapiesthat can promote native AVF maturation and prevent vascularstenosis, because these are the major problems that are associatedwith hemodialysis vascular access dysfunction. Dialysis graftsand fistulae are ideally suited to locally delivered therapiesin view of their superficial anatomic location and the prospectof repeat delivery of local agents at the time of hemodialysisthrough the dialysis needles. Particular emphasis thereforeis placed on these approaches. The potential therapies discussedspan the entire spectrum of drug and nonpharmacologic interventions,from agents and techniques that are currently available on themarket to promising therapies that are still in the early experimentalstage.
Altering Upstream Events
Although the potential benefit from manipulation of upstreamhemodynamic stressors could be very significant, this has beena relatively neglected area of research and innovation. Numerousexperimental studies have demonstrated clearly that turbulent,low-flow, and low-shear-stress systems predispose to neointimalhyperplasia (21, 56) and that reversal of such abnormalitiescan reduce the degree of neointimal hyperplasia (57). Interventionsthat aim to prevent the development of low shear stress thereforemay result in less neointimal hyperplasia. At a clinical level,however, the only such intervention in the setting of hemodialysisvascular access has been studies that have used the Venaflohooded graft with initial studies suggesting improved survival(58). The only other "upstream" intervention that has been shownto be successful in enhancing fistula maturation is the ligationof accessory veins (59–62).
Currently Available Systemic Agents
There has been a lot of interest in the use of currently availableagents that have the potential to block smooth muscle cell proliferationand migration and/or thrombosis in the setting of hemodialysisvascular access dysfunction. Unfortunately, much of these dataare anecdotal and involve a very small number of patients (63).Dipyridamole (64) and fish oil (65) have been shown to preventstenosis and thrombosis in PTFE dialysis grafts (primary prevention)in randomized clinical trials, whereas angiotensin-convertingenzyme inhibitors have been shown to be of benefit in retrospectiveregistry analyses (66). On the basis of some of these data,the National Institutes of Health–sponsored Dialysis AccessConsortium is conducting two large, multicenter, randomized,prospective, primary prevention studies (Aggrenox in dialysisgrafts and clopidogrel in native fistulae). Two newer agentsthat have been shown to have potent inhibitory effects on vascularstenosis in experimental models are sirolimus (67) and the peroxisomeproliferator–activated receptor agonist rosiglitazone(68). In addition to blocking smooth muscle cell migration andproliferation, both sirolimus and rosiglitazone seem to modulatethe relative number of smooth muscle and endothelial cell progenitorsin experimental models (45, 69), emphasizing the importanceof these alternative mechanisms in the pathogenesis of neointimalhyperplasia. Preliminary clinical studies demonstrate a reductionin in-stent restenosis after coronary angioplasty, with theoral administration of sirolimus (70) and rosiglitazone (71).The role of these agents in the clinical setting of hemodialysisvascular access dysfunction is unknown.
Radiation Therapy
The rationale for the use of radiation therapy for the preventionand treatment of vascular stenosis as a result of neointimalhyperplasia stems from (1) the potent in vitro antiproliferativeeffect of radiation therapy on smooth muscle cells, endothelialcells, and macrophages (72–74) and (2) possible beneficialeffects on vascular remodeling (vessel wall dilation) (75).Experimental angioplasty models have demonstrated significantreductions in luminal stenosis with endovascular and externalbeam radiation therapy (76, 77), and these findings have beenconfirmed in large clinical studies of catheter-based radiationtherapy for coronary restenosis (78, 79). Both external beamand endovascular radiation therapy have been shown to reducevascular stenosis in experimental arteriovenous graft (80, 81)and native fistula (82) models. In addition, a pilot study ofendovascular radiation therapy after angioplasty (BRAVO-I) indialysis patients with patent but dysfunctional grafts documentedan improvement in 6-mo target lesion primary patency (thrombosisor angioplasty of the irradiated lesion) in the radiation group(83, 84). Unfortunately, this improvement in target lesion primarypatency did not translate into better cumulative patency (overallgraft survival regardless of the number of thrombotic episodes/angioplasties)at either 6 or 12 mo after intervention. Some data from a largerstudy of endovascular radiation therapy in thrombosed PTFE dialysisgrafts should be available shortly (BRAVO-II).
Gene Therapy
Gene therapy could become an effective means of local therapyfor neointimal hyperplasia in dialysis grafts and fistulae (85),especially if improvements continue to be made in the safetyand the efficacy of delivery techniques (86). Currently, inhibitionof neointimal hyperplasia in experimental angioplasty modelshas been achieved by the gene transfer of endothelial and inducibleNO synthase, cyclin-dependent kinase inhibitors, retinoblastomaprotein, hepatocyte growth factor, and transcription factorssuch as Edifoligide (E2F) (24). A phase II trial of the E2Fdecoy is currently in progress in arteriovenous grafts, butit is unlikely that this will progress into a larger study inview of the lack of efficacy of E2F decoys in the setting ofsaphenous vein bypass grafting in the coronary circulation (PREVENTIV) (87) and in peripheral vein bypass grafting (PREVENT III)(88).
Local Drug Delivery Stent Adaptations.
The main advantage of stent placement after vascular injury(angioplasty) is a reduction in adverse vascular remodeling(see Importance of Vascular Remodeling section, above). Unfortunately,bare-metal stents are prone to develop an aggressive in-stentrestenosis. Recent experimental (89, 90) and clinical studies(37, 38) with placement of drug-eluting stents that releasepaclitaxel and sirolimus (both are small-molecule inhibitorsof cell-cycle progression) after coronary angioplasty have showna marked reduction in in-stent restenosis as compared with bare-metalstents (Figure 11). Most impressive, this effect seems to belong lasting (up to 4 yr) (91). A recent direct comparison (SIRTAXstudy) between the sirolimus-eluting (Cypher; Cordis, MiamiLakes, FL) and paclitaxel-eluting (TAXUS; Boston Scientific,Natick, MA) stents suggested that the sirolimus-eluting stentis superior with regard to target lesion revascularization andlate lumen loss (92).
Figure 11. Drug-eluting stents. Sirolimus and paclitaxel are small-molecule inhibitors of cell-cycle proliferation (a, c, and e). When loaded into a polymer and coated onto coronary stents (d), they result in a significant inhibition of in-segment restenosis (b and f). Adapted from (37, 38).
In the setting of dialysis access dysfunction, there does notseem to be any benefit from the routine placement of bare-metalstents after angioplasty (93, 94). There is no information onthe placement of drug-eluting stents in patients with dialysisgrafts and fistulae, although a recent study in a pig modelof arteriovenous graft stenosis demonstrated an improvementin luminal stenosis at 28 d when sirolimus-eluting stents, ascompared with bare-metal stents, were placed at the time ofsurgery (95). One potential disadvantage of drug-eluting stentsin the hemodialysis population could be an increased risk forbleeding if both aspirin and clopidogrel are required afterstent placement (96).
Local Perivascular Drug Delivery.
The theoretical advantages of perivascular drug delivery forneointimal hyperplasia are as follows. (1) Application of thedrug of choice directly to the adventitia ("outside-in approach")may be far more effective in blocking adventitial activationand fibroblast migration as compared with local endovascularapplication through a drug-eluting stent ("inside-out approach").This could be of great clinical relevance in view of the importanceof the adventitia in the pathogenesis of neointimal hyperplasia(discussed in Pathology and Pathogenesis of Native Fistula andPTFE Graft Failure section, above). (2) Delivery of the therapeuticagent to the adventitia results in a gradient of drug concentrations,with the highest levels in the adventitial layer and the lowestlevels at the endothelial layer. In the specific context ofantiproliferative agents (the most common type of drug therapyfor neointimal hyperplasia), lower drug concentrations at theendothelial layer actually could be beneficial by allowing endothelializationof the region of vascular injury. (3) The direct local applicationof a small amount of drug can result in high concentrationsat the site of vascular injury (neointimal hyperplasia) withminimal systemic toxicity (as a result of the small dose used).Most important, the validity of such an approach has been documentedin a number of experimental angioplasty models using agentssuch as NO, paclitaxel, and the tyrphostins (24).
In most cases, perivascular delivery systems involve the useof polymer-based systems that are not drug specific. A varietyof simple small-molecule drugs, antibodies, or nucleotides canbe incorporated into the polymers and delivered locally (seeexamples above). Some polymers, such as a triblock co-polymerof polylactide-polyethyleneglycol-polylactide, exhibit a temperature-dependentreversible transition between solution and gel phases. The co-polymerdissolves in water to form a free-flowing liquid at low temperaturesbut turns into a water-insoluble gel within seconds at bodytemperature. Therefore, these polymers can be mixed with drugsin the aqueous phase and injected perivascularly, whereby theyform a gel depot at the desired location for sustained drugrelease. In addition, both in vitro (97) and in vivo (98) pharmacokineticstudies in the porcine model have demonstrated that drugs suchas dipyridamole that are delivered by these polymers perivascularlydiffuse through arteries, veins, and PTFE grafts and thereforeare available throughout the thickness of the vessel for theinhibition of smooth muscle cells or myofibroblasts.
Paclitaxel has been incorporated into these thermosensitivepolymers, injected into the perivascular area of hemodialysisgrafts several weeks after graft placement, and shown to inhibitneointimal hyperplasia in a canine model (32). An alternativeapproach that involves the placement of paclitaxel-loaded wrapsaround the graft–vein anastomosis also has been shownto be effective in reducing neointimal hyperplasia and luminalstenosis (99). In both instances, the polymer in which the drugis trapped provides a mean of sustained delivery, at the anastomosesbetween the native vessels and the graft, allowing the gradualrelease of the drug over weeks or months. Theoretically, repeatedpercutaneous injection of antiproliferative drugs at varioustime intervals using these polymers may provide sufficientlyhigh local concentrations for months to years for the preventionof graft stenosis.
In addition, the release profile of drugs from thermosensitivepolymeric gels can be manipulated further by combining the thermosensitivegels with a variety of drug platforms such as poly(lactide-co-glycolide)microspheres (100). Using dipyridamole as the test molecule,it can be demonstrated clearly that the combination of thermosensitivegel with dipyridamole encapsulated in microspheres further delaysand decreases the rate of release of this drug. The delay anddecrease in release rate can be manipulated by altering themolecular weights of the polymers that are used in the preparationof the microspheres. Therefore, slow or fast release rates atdifferent time points after injection of the perivascular polymerscan be achieved, and this then can be titrated so that the drugrelease profile can have a maximal impact on the known temporalcourse of arteriovenous graft stenosis.
We believe that perivascular drug delivery using a variety ofpolymer combinations for the sustained delivery of antiproliferativedrugs and perhaps antiplatelet agents is an intriguing and promisingapproach to the clinical problem of hemodialysis vascular accessdysfunction, which needs further development. It is likely thatdialysis access grafts and fistulae could be the ideal clinicalmodel for perivascular polymer-based drug delivery systems inview of their superficial position, their remoteness from vitalorgans, and the potential logistic benefits of easy repeatedaccess to the patient during the hemodialysis session.
Utilizing PTFE Graft as a Conduit for Drug Delivery.
Despite their increased rates of stenosis and thrombosis, PTFEdialysis access grafts may have an important advantage overnative AVF in that the graft material itself could be used asa platform for local drug delivery to prevent infection, thrombosis,and stenosis. In view of the ability of NO to stabilize endothelialcells and also block smooth muscle cell activation, Frost etal. (101) devised a number of different polymers loaded withNO-releasing substances that can be coated onto PTFE graft material.Pilot studies using these NO-releasing grafts have demonstrateda reduction in thrombosis and an improvement in patency, ina sheep model of arteriovenous graft stenosis (102). A similarapproach with other antiproliferative and antithrombotic agentsneeds to be pursued aggressively.
Endothelial Seeding and Manipulation of Progenitor Bone Marrow–Derived Cells
The holy grail of experimental vascular surgery is the dreamof being able to seed vascular grafts with a layer of endothelialcells that have the capacity to release endogenous inhibitorsof stenosis and thrombosis such as NO and prostacyclin. Unfortunately,this has proved to be extremely challenging, although some encouragingresults have been obtained through the use of shear stress activationof endothelial cells (103) and electrostatic seeding (104).It is interesting that endothelial progenitor cells seem tohave a far greater propensity (as compared with differentiatedendothelial cells) to attach to prosthetic stent and graft material,and the best results to date have been obtained by plating exvivo cultures of EPC onto grafts and stents (Figure 12, a and b)(105, 106). Unfortunately, this is a time-consuming and technicallydemanding process that may not be suited to clinical use. Anelegant alternative approach has been the development of vasculargrafts or stents that are coated with an antibody to CD34, whichis an antigen that is present on EPC. As expected, when thesestents are placed within the vascular tree, they are able toattract in EPC and undergo rapid endothelialization, which maytranslate into a reduction in neointimal hyperplasia (53, 107).The first clinical studies with these CD34 antibody–coatedstents demonstrate both safety and feasibility (108). However,a recent study demonstrates increased neointimal hyperplasiain CD34-coated arteriovenous grafts, despite an increase inendothelialization (109).
Figure 12. EPC can seed prosthetic stents and grafts. Ex vivo EPC cultures can successfully coat prosthetic stents with an endothelial cell layer. (a) Low-power scanning electron microscopy picture of a prosthetic stent coated with EPC (bar = 1 mm). (b) Confocal scanning laser microscopy picture of EPC stained with PicoGreen fluorescence, on an expanded stent (bar = 1 mm). Coating prosthetic stents with an antibody to the EPC marker CD34 can result in rapid endothelialization of prosthetic stents. Note the large number of cells seen on low-power scanning electron microscopy at 1 h on the CD34-coated stents (d) as compared with the noncoated stents (c). Magnifications: x35 in a, x30 in b. Adapted from references (105) (a and b) and (53) and Dr. M. Kutryk (personal communication) (c and d).
Another approach that is completely experimental at presentis optimizing the balance between neointima-enhancing smoothmuscle progenitor cells and neointima-attenuating EPC. Thispotentially could be done by modulating integrin expressionon smooth muscle progenitor cells (110) to reduce their adhesionto sites of vascular injury while enhancing the mobilizationof EPC from the bone marrow to sites of vascular injury (48).
Promoting Native AVF Maturation
As the rate of AVF placement increases, a corresponding increasein the incidence of primary nonfunction (failure to mature)is expected. In marked contrast to the many therapies that havepromise in reducing neointimal hyperplasia, very few novel therapiesfocus on enhancing fistula maturation. Part of the reason forthis is that primary nonfunction is a poorly understood, multifactorialprocess that is unlikely to respond to single-drug therapy.An exciting alternative approach (cell-based therapy) has beenpioneered by Nugent et al. (33, 111); it involves embeddingspecific cell types (e.g., endothelial cells) into a perivascularGelfoam cuff. These cells then could produce an array of mediatorsthat hopefully would mimic the two main functions of endothelialcells (promote vascular dilation and inhibit neointimal hyperplasia).At a practical level, however, numerous clinical studies havebeen able to demonstrate a significant improvement in the maturationof AVF through an aggressive preoperative ultrasound evaluationof arterial and venous diameters (112–114). Malovrh etal. (113), for example, demonstrated an immediate patency rateof 92% in patients with a preoperative internal diameter of>1.5 mm in the feeding artery as compared with a maturationrate of 45% in patients with an internal diameter of <1.5mm. At 12 wk, the patency rates in the two groups were 83 and36%, respectively.
Other Therapies Photodynamic Therapy.
The rationale behind photodynamic therapy is to administer aphotosensitizer, either systemically or locally, and then exposethe region of vascular injury to light radiation (external orendovascular), resulting in the production of singlet oxygenthat can cause cell death in the region that is exposed to lightradiation. Photodynamic therapy has been used in experimentalmodels of peripheral and coronary angioplasty (115, 116) andin an arteriovenous graft model (Miravant Medical Technologies,Santa Barbara, CA) (117) and has demonstrated safety, feasibility,and preliminary efficacy. Some anecdotal clinical data alsoare available, but a large phase III study in a clinical modelof vascular injury is lacking.
Cryoplasty.
Cryoplasty is the combination of a standard angioplasty withthe simultaneous delivery of cold thermal energy (–10°Cfor 60 s; CryoVascular Systems, Los Gatos, CA). Initial clinicalstudies in femoropopliteal angioplasty (15 patients) (118) andrapidly recurrent restenosis in arteriovenous dialysis grafts(five patients) (119) suggest safety, feasibility, and preliminaryefficacy.
Stent Grafts.
In contrast to the lack of data on the use of drug-eluting stentsin dialysis access dysfunction and the lack of efficacy of bare-metalstents after angioplasty, a recent prospective, randomized studycompared the placement of stent grafts (Bard Peripheral Vascular,Tempe, AZ) after angioplasty with angioplasty alone in PTFEgraft stenosis. Placement of stent grafts resulted in a significantimprovement in primary patency at 6 mo (53 versus 29%) (120).This is an important result that could have immediate clinicalapplicability.
We are in the midst of fundamental changes in the way we addressthe clinical problem of hemodialysis vascular access dysfunction.The purpose of this last section is both to summarize wherewe stand and to make suggestions for future scientific advancein this field.
There is a clear appreciation of the magnitude of the clinicalproblem and the understanding that prevention rather than mechanicaltreatment should be the goal in reducing the morbidity thatis associated with hemodialysis arteriovenous access dysfunction.In particular, we now recognize that all vascular manipulations(surgery or balloon angioplasty) cause endothelial and smoothmuscle cell injury, which results in a restenotic process. Therefore,these interventions need to be linked to therapies that cantarget both the traditional and the alternative pathways thatare involved in the pathogenesis of neointimal hyperplasia andvascular stenosis (see below).
We have a reasonable understanding of the pathology and thepathogenesis of venous neointimal hyperplasia and vascular stenosisin the setting of dialysis access dysfunction. However, thepast few years have seen great advances in the elucidation ofalternative mechanisms that are responsible for neointimal hyperplasia,and we need to assess the impact of these new paradigms on theclinical problem of hemodialysis vascular access dysfunction.
The combination of advances in cellular and molecular pathobiology,biomaterials, and drug delivery techniques has resulted in manyinnovative therapies for neointimal hyperplasia. We need toidentify (from this plethora of interventions) the therapiesthat are best suited for clinical use in the setting of hemodialysisvascular access dysfunction; we need to test these therapiesin the excellent animal models of arteriovenous stenosis thatare available to us, and we need to move the most promisingof these therapies from the laboratory to the clinic.
In parallel, clinical trials need to be conducted to examineboth the currently available oral agents that are known to inhibitsmooth muscle cell proliferation or migration and that havealready shown clinical efficacy in the arterial neointimal settingand to test therapies that target novel pathways of neointimalhyperplasia and vascular remodeling. In testing new therapies,however, we need to recognize the redundancy and the pleiotropismof biologic systems and appreciate that combination therapymay yield the greatest clinical benefits (this holds true forboth experimental and clinical studies).
To optimize the use of resources, we need to standardize clinicaltrial end points in this field for both primary and secondaryprevention studies (some guidelines are already available [121,122]) and identify population subsets (patients who are proneto early and aggressive stenosis) so that appropriate stratificationcan be performed.
Finally we need to emphasize that dialysis grafts and fistulaecould be the ideal clinical model for testing new locally deliveredtherapies for neointimal hyperplasia, in view of their superficiallocation, the aggressive nature of the lesion, and the frequentfollow-up of these patients in the dialysis clinics. The resultsfrom such trials then could be applied to other clinical conditionsthat are characterized by neointimal hyperplasia, such as postangioplastyrestenosis, peripheral vascular disease, and coronary arterybypass graft stenosis.
Acknowledgments
This work was supported by the Norman Coplon Extramural GrantProgram (Satellite Dialysis), the Dialysis Research Foundation,the National Kidney Foundation of Utah and Idaho, the NationalInstitutes of Health (NIH RO1DK61689 and RO1HL67646), and theMerit Review Program of the Department of Veterans Affairs.
Footnotes
Published online ahead of print. Publication date availableat www.jasn.org
United States Renal Data System:
2002 Annual Data Report, Bethesda, National Institutes of Health, National Institute of Diabetes and Digestive Diseases, 2002
Egger P:
Trends in Medicare Expenditure for Vascular Access. Presented at the Cincinnati Hemodialysis Vascular Access Symposium, March 19, 2004
Schwab SJ, Harrington JT, Singh A, Roher R, Shohaib S, Perrone R, Meyer K, Beasley D: Vascular access for hemodialysis.
Kidney Int 55
: 2078
–2090, 1999[CrossRef][Medline]
Miller PE, Tolwani A, Luscy CP, Deierhoi MH, Bailey R, Redden DT, Allon M: Predictors of adequacy of arteriovenous fistulas in hemodialysis patients.
Kidney Int 56
: 275
–280, 1999[CrossRef][Medline]
Leapman SB, Boyle M, Pescovitz MD, Milgrom ML, Jindal RM, Filo RS: The arteriovenous fistula for hemodialysis access: Gold standard or archaic relic?
Am Surg 62
: 652
–656; discussion 656–657, 1996[Medline]
Miller PE, Carlton D, Deierhoi MH, Redden DT, Allon M: Natural history of arteriovenous grafts in hemodialysis patients.
Am J Kidney Dis 36
: 68
–74, 2000[Medline]
Turmel-Rodrigues L, Pengloan J, Baudin S, Testou D, Abaza M, Dahdah G, Mouton A, Blanchard D: Treatment of stenosis and thrombosis in haemodialysis fistulas and grafts by interventional radiology.
Nephrol Dial Transplant 15
: 2029
–2036, 2000[Abstract/Free Full Text]
Asif A, Gadalean FN, Merrill D, Cherla G, Cipleu CD, Epstein DL, Roth D: Inflow stenosis in arteriovenous fistulas and grafts: A multicenter, prospective study.
Kidney Int 67
: 1986
–1992, 2005[CrossRef][Medline]
Stracke S, Konner K, Kostlin I, Friedl R, Jehle PM, Hombach V, Keller F, Waltenberger J: Increased expression of TGF-beta1 and IGF-I in inflammatory stenotic lesions of hemodialysis fistulas.
Kidney Int 61
: 1011
–1019, 2002[CrossRef][Medline]
Weiss MF, Scivittaro V, Anderson JM: Oxidative stress and increased expression of growth factors in lesions of failed hemodialysis access.
Am J Kidney Dis 37
: 970
–980, 2001[Medline]
Beathard GA: The treatment of vascular access dysfunction: A nephrologists view and experience.
Adv Ren Replace Ther 1
: 131
–147, 1994[Medline]
Roy-Chaudhury P, Kelly BS, Miller MA, Reaves A, Armstrong J, Nanayakkara N, Heffelfinger SC: Venous neointimal hyperplasia in polytetrafluoroethylene dialysis grafts.
Kidney Int 59
: 2325
–2334, 2001[Medline]
Rekhter MD, Gordon D: Active proliferation of different cell types, including lymphocytes, in human atherosclerotic plaques.
Am J Pathol 147
: 668
–677, 1995[Abstract]
Swedberg SH, Brown BG, Sigley R, Wight TN, Gordon D, Nicholls SC: Intimal fibromuscular hyperplasia at the venous anastomosis of PTFE grafts in hemodialysis patients. Clinical, immunocytochemical, light and electron microscopic assessment.
Circulation 80
: 1726
–1736, 1989[Abstract/Free Full Text]
Heine GH, Ulrich C, Kohler H, Girndt M: Is AV fistula patency associated with angiotensin-converting enzyme (ACE) polymorphism and ACE inhibitor intake?
Am J Nephrol 24
: 461
–468, 2004[CrossRef][Medline]
Allon M, Robbin ML: Increasing arteriovenous fistulas in hemodialysis patients: Problems and solutions.
Kidney Int 62
: 1109
–1124, 2002[CrossRef][Medline]
Remuzzi A, Ene-Iordache B, Mosconi L, Bruno S, Anghileri A, Antiga L, Remuzzi G: Radial artery wall shear stress evaluation in patients with arteriovenous fistula for hemodialysis access.
Biorheology 40
: 423
–430, 2003[Medline]
Paszkowiak JJ, Dardik A: Arterial wall shear stress: Observations from the bench to the bedside.
Vasc Endovascular Surg 37
: 47
–57, 2003[Abstract/Free Full Text]
Mezzano D, Pais EO, Aranda E, Panes O, Downey P, Ortiz M, Tagle R, Gonzalez F, Quiroga T, Caceres MS, Leighton F, Pereira J: Inflammation, not hyperhomocysteinemia, is related to oxidative stress and hemostatic and endothelial dysfunction in uremia.
Kidney Int 60
: 1844
–1850, 2001[CrossRef][Medline]
Chang CJ, Ko PJ, Hsu LA, Ko YS, Ko YL, Chen CF, Huang CC, Hsu TS, Lee YS, Pang JH: Highly increased cell proliferation activity in the restenotic hemodialysis vascular access after percutaneous transluminal angioplasty: Implication in prevention of restenosis.
Am J Kidney Dis 43
: 74
–84, 2004[CrossRef][Medline]
Roy-Chaudhury P, Kelly BS, Zhang J, Narayana A, Desai P, Melham M, Duncan H, Heffelfinger SC: Hemodialysis vascular access dysfunction: From pathophysiology to novel therapies.
Blood Purif 21
: 99
–110, 2003[CrossRef][Medline]
Adams LD, Geary RL, McManus B, Schwartz SM: A comparison of aorta and vena cava medial message expression by cDNA array analysis identifies a set of 68 consistently differentially expressed genes, all in aortic media.
Circ Res 87
: 623
–631, 2000[Abstract/Free Full Text]
Kim SJ, Masaki T, Leypoldt JK, Kamerath CD, Mohammad SF, Cheung AK: Arterial and venous smooth-muscle cells differ in their responses to antiproliferative drugs.
J Lab Clin Med 144
: 156
–162, 2004[CrossRef][Medline]
Kim SJ, Masaki T, Rowley R, Kamerath CD, Leypoldt JK, Mohammed SF, Cheung AK: Effect of radiation on arterial and venous smooth muscle cells in culture [Abstract].
J Am Soc Nephrol 12
: 292A
, 2001
Kelly BS, Heffelfinger SC, Whiting JF, Miller MA, Reaves A, Armstrong J, Narayana A, Roy-Chaudhury P: Aggressive venous neointimal hyperplasia in a pig model of arteriovenous graft stenosis.
Kidney Int 62
: 2272
–2280, 2002[CrossRef][Medline]
Kohler TR, Kirkman TR: Dialysis access failure: A sheep model of rapid stenosis.
J Vasc Surg 30
: 744
–751, 1999[CrossRef][Medline]
Masaki T, Rathi R, Zentner G, Leypoldt JK, Mohammad SF, Burns GL, Li L, Zhuplatov S, Chirananthavat T, Kim SJ, Kern S, Holman J, Kim SW, Cheung AK: Inhibition of neointimal hyperplasia in vascular grafts by sustained perivascular delivery of paclitaxel.
Kidney Int 66
: 2061
–2069, 2004[CrossRef][Medline]
Nugent HM, Groothuis A, Seifert P, Guerraro JL, Nedelman M, Mohanakumar T, Edelman ER: Perivascular endothelial implants inhibit intimal hyperplasia in a model of arteriovenous fistulae: A safety and efficacy study in the pig.
J Vasc Res 39
: 524
–533, 2002[CrossRef][Medline]
Libby P, Tanaka H: The molecular bases of restenosis.
Prog Cardiovasc Dis 40
: 97
–106, 1997[CrossRef][Medline]
O’Brien JE Jr, Shi Y, Fard A, Bauer T, Zalewski A, Mannion JD: Wound healing around and within saphenous vein bypass grafts.
J Thorac Cardiovasc Surg 114
: 38
–45, 1997[Abstract/Free Full Text]
Morice MC, Serruys PW, Sousa JE, Fajadet J, Ban Hayashi E, Perin M, Colombo A, Schuler G, Barragan P, Guagliumi G, Molnar F, Falotico R: A randomized comparison of a sirolimus-eluting stent with a standard stent for coronary revascularization.
N Engl J Med 346
: 1773
–1780, 2002[Abstract/Free Full Text]
Stone GW, Ellis SG, Cox DA, Hermiller J, O’Shaughnessy C, Mann JT, Turco M, Caputo R, Bergin P, Greenberg J, Popma JJ, Russell ME: A polymer-based, paclitaxel-eluting stent in patients with coronary artery disease.
N Engl J Med 350
: 221
–231, 2004[Abstract/Free Full Text]
Shi Y, O’Brien JE, Fard A, Mannion JD, Wang D, Zalewski A: Adventitial myofibroblasts contribute to neointimal formation in injured porcine coronary arteries.
Circulation 94
: 1655
–1664, 1996[Abstract/Free Full Text]
Shi Y, O’Brien JE Jr, Mannion JD, Morrison RC, Chung W, Fard A, Zalewski A: Remodeling of autologous saphenous vein grafts. The role of perivascular myofibroblasts.
Circulation 95
: 2684
–2693, 1997[Abstract/Free Full Text]
Scott NA, Cipolla GD, Ross CE, Dunn B, Martin FH, Simonet L, Wilcox JN: Identification of a potential role for the adventitia in vascular lesion formation after balloon overstretch injury of porcine coronary arteries.
Circulation 93
: 2178
–2187, 1996[Abstract/Free Full Text]
Roy-Chaudhury P, McKee L, Miller M, Reaves A, Armstrong J, Duncan H, Munda R, Kelly B, Heffelfinger S: Adventitial fibroblasts contribute to venous neointimal hyperplasia in PTFE dialysis grafts [Abstract].
J Am Soc Nephrol 12
: 301A
, 2001[CrossRef]
Sata M, Saiura A, Kunisato A, Tojo A, Okada S, Tokuhisa T, Hirai H, Makuuchi M, Hirata Y, Nagai R: Hematopoietic stem cells differentiate into vascular cells that participate in the pathogenesis of atherosclerosis.
Nat Med 8
: 403
–409, 2002[CrossRef][Medline]
Diao YGM, Guthrie S, Scott E, Segal MS: Role of circulating progenitor cells in the development of venous intimal hyperplasia [Abstract].
J Am Soc Nephrol 13
: 58A
–59A, 2002[CrossRef]
Fukuda D, Sata M, Tanaka K, Nagai R: Potent inhibitory effect of sirolimus on circulating vascular progenitor cells.
Circulation 111
: 926
–931, 2005[Abstract/Free Full Text]
Urbich C, Dimmeler S: Endothelial progenitor cells: Characterization and role in vascular biology.
Circ Res 95
: 343
–353, 2004[Abstract/Free Full Text]
Asahara T, Murohara T, Sullivan A, Silver M, van der Zee R, Li T, Witzenbichler B, Schatteman G, Isner JM: Isolation of putative progenitor endothelial cells for angiogenesis.
Science 275
: 964
–967, 1997[Abstract/Free Full Text]
Szmitko PE, Fedak PW, Weisel RD, Stewart DJ, Kutryk MJ, Verma S: Endothelial progenitor cells: New hope for a broken heart.
Circulation 107
: 3093
–3100, 2003[Free Full Text]
Urao N, Azuma A: Erythropoietin prevents neointimal hyperplasia alter intimal injury by mobilizing endothelial precursor cells to accelerate endothelialization [Abstract 699]. Presented at the American Heart Association Annual Meeting, New Orleans, November 2004
Werner N, Junk S, Laufs U, Link A, Walenta K, Bohm M, Nickenig G: Intravenous transfusion of endothelial progenitor cells reduces neointima formation after vascular injury.
Circ Res 93
: e17
–e24, 2003[Abstract/Free Full Text]
Griese DP, Ehsan A, Melo LG, Kong D, Zhang L, Mann MJ, Pratt RE, Mulligan RC, Dzau VJ: Isolation and transplantation of autologous circulating endothelial cells into denuded vessels and prosthetic grafts: Implications for cell-based vascular therapy.
Circulation 108
: 2710
–2715, 2003[Abstract/Free Full Text]
Shears LL 2nd, Kibbe MR, Murdock AD, Billiar TR, Lizonova A, Kovesdi I, Watkins SC, Tzeng E: Efficient inhibition of intimal hyperplasia by adenovirus-mediated inducible nitric oxide synthase gene transfer to rats and pigs in vivo.
J Am Coll Surg 187
: 295
–306, 1998[CrossRef][Medline]
Kipshidze N, Dangas G, Tsapenko M, Moses J, Leon MB, Kutryk M, Serruys P: Role of the endothelium in modulating neointimal formation: Vasculoprotective approaches to attenuate restenosis after percutaneous coronary interventions.
J Am Coll Cardiol 44
: 733
–739, 2004[Abstract/Free Full Text]
Garg UC, Hassid A: Nitric oxide-generating vasodilators and 8-bromo-cyclic guanosine monophosphate inhibit mitogenesis and proliferation of cultured rat vascular smooth muscle cells.
J Clin Invest 83
: 1774
–1777, 1998
Kong D, Melo LG, Gnecchi M, Zhang L, Mostoslavsky G, Liew CC, Pratt RE, Dzau VJ: Cytokine-induced mobilization of circulating endothelial progenitor cells enhances repair of injured arteries.
Circulation 110
: 2039
–2046, 2004[Abstract/Free Full Text]
Corpataux JM, Haesler E, Silacci P, Ris HB, Hayoz D: Low-pressure environment and remodelling of the forearm vein in Brescia-Cimino haemodialysis access.
Nephrol Dial Transplant 17
: 1057
–1062, 2002[Abstract/Free Full Text]
Top
Abstract
Introduction and Types of...
(the last has been shown to co-localize with markers of oxidative stress) within the media and intima (12, 13). 
); chemokines (monocyte chemoattractant protein-1 and RANTES); vasoactive molecules (nitric oxide [NO] and endothelin); adhesion molecules (intercellular adhesion molecule-1 and P-selectin); and molecules such as osteopontin, apolipoprotein E, matrix metalloproteinase-2, and human hepatocyte growth factor (24). It should be emphasized, however, that most of this information is derived from arterial angioplasty models rather than from vascular (venous) anastomosis models. 
agonist rosiglitazone (68). In addition to blocking smooth muscle cell migration and proliferation, both sirolimus and rosiglitazone seem to modulate the relative number of smooth muscle and endothelial cell progenitors in experimental models (45, 69), emphasizing the importance of these alternative mechanisms in the pathogenesis of neointimal hyperplasia. Preliminary clinical studies demonstrate a reduction in in-stent restenosis after coronary angioplasty, with the oral administration of sirolimus (70) and rosiglitazone (71). The role of these agents in the clinical setting of hemodialysis vascular access dysfunction is unknown. 
