Version May 2024
INTRODUCTION — Hemodialysis requires access to blood vessels capable of providing rapid extracorporeal blood flow. These requirements are currently best met by arteriovenous (AV) access. Although AV fistulas are preferred for hemodialysis, AV grafts are sometimes necessary but have higher failure rates. Failure can be related to stenotic lesions affecting the feeding arteries, within the graft or in the draining veins, including the central veins. AV graft failure can also be related to complications such as pseudoaneurysm or other conditions that lead to sacrifice of the graft.
The treatment of AV graft failure due to venous stenosis and thrombotic complications are reviewed here. Issues surrounding clinical monitoring and surveillance of AV grafts to prevent thrombosis are presented in detail separately. (See "Physical examination of the arteriovenous graft" and "Clinical monitoring and surveillance of hemodialysis arteriovenous grafts to prevent thrombosis".)
The management of stenosis and thrombosis of hemodialysis AV fistulas is discussed elsewhere. (See "Early evaluation of the newly created hemodialysis arteriovenous fistula" and "Failure of the mature hemodialysis arteriovenous fistula".)
AV GRAFT DYSFUNCTION AND FAILURE
Incidence — When stenotic lesions (primarily neointimal hyperplasia) develop in association with an AV graft, it often results in thrombosis. More than 90 percent of thrombosed AV grafts have a stenotic lesion, suggesting that such an anatomic abnormality is required for AV graft thrombosis. The majority of AV grafts develop stenosis or thrombosis. Among 649 patients with new AV grafts enrolled in the Dialysis Access Consortium (DAC) Study, 77 percent developed stenosis or thrombosis within the first year [1]. Similarly, among 201 patients with a new AV graft enrolled in the Fish Oil Inhibition of Stenosis in Hemodialysis Grafts (FISH) Study, 62 percent developed stenosis or thrombosis within one year [2].
Etiology
Immediate — Immediate failure of AV grafts following their creation is usually due to technical issues, often related to creation of an inadequate anastomosis as a result of severe arterial calcification. A large, single-center study observed immediate failure in 5.8 percent of upper extremity AV grafts and 12.7 percent of thigh grafts [3]. (See "Arteriovenous graft creation for hemodialysis and its complications".)
Delayed — AV graft failure beyond the postoperative period (delayed graft failure) is typically related to stenotic vascular lesions, most commonly affecting the venous outflow anastomosis due to neointimal hyperplasia. Critical stenosis can lead to sudden AV graft thrombosis. AV grafts less commonly fail due to infection. In a large single-center observational study of 310 AV grafts, 80 percent of graft failures were due to thrombosis, 14 percent were due to infection, and 3 percent were due to a large pseudoaneurysm with impending rupture. The remainder (2 percent) were caused by intractable central vein stenosis requiring AV graft ligation, and distal ischemia (steal syndrome) [4]. (See 'Treatment of stenosis' below and "Endovascular intervention for the treatment of stenosis in the arteriovenous access", section on 'Lesions unique to AV grafts' and "Arteriovenous graft creation for hemodialysis and its complications" and "Arteriovenous fistula creation for hemodialysis and its complications".)
Some patients with thrombosis have an underlying hypercoagulable state as a cause for AV graft thrombosis [5-8]. (See 'Hypercoagulable evaluation' below.)
CLINICAL PRESENTATION AND EVALUATION — Most cases of AV graft thrombosis are preceded by rapidly progressive stenosis at the vein-graft anastomosis that can be documented by access surveillance or clinical monitoring. However, up to 25 percent of AV grafts clot fairly abruptly without prior indication of critical AV graft stenosis. (See "Physical examination of the arteriovenous graft" and "Clinical monitoring and surveillance of hemodialysis arteriovenous grafts to prevent thrombosis".)
Abnormalities on clinical monitoring — For patients with an abnormal clinical examination or abnormalities on monitoring and surveillance, the cause of the problem may be obvious with physical examination [9].
In one review, clinical monitoring tools included abnormal physical examination (absent thrill, abnormal graft auscultation, or edema distal to the graft); abnormalities related to the dialysis session (difficulty in cannulation, aspiration of clots, inability to achieve the target dialysis blood flow, prolonged bleeding from the needle sites); and an unexplained, sustained fall (>0.2 units) in delivered Kt/V, despite following a constant dialysis prescription [10]. (See "Prescribing and assessing adequate hemodialysis", section on 'Kt/V'.)
Detection of one of these abnormalities led to routine angiograms, and, in 69 percent of cases, a lesion requiring angioplasty was detected. In a single-center randomized trial, the treatment arm undergoing clinical monitoring without ultrasound surveillance underwent 76 angioplasty procedures [11]. These were prompted by an abnormality in physical examination in 21 percent, abnormalities related to the dialysis session in 49 percent, and an unexplained decrease of Kt/V in 30 percent. Even in the group undergoing duplex ultrasound surveillance, 24 percent of the angioplasties were still prompted by abnormalities that arose during clinical monitoring.
Evaluation of stenotic lesions — In agreement with the Kidney Disease Outcomes Quality Initiative (KDOQI) guidelines, we define a clinically significant stenosis in an AV graft as a greater than 50 percent diameter narrowing in conjunction with the abnormal clinical findings, such as decreasing intragraft blood flow (less than 600 mL/min) or elevated static pressure within the graft [12,13]. (See "Clinical monitoring and surveillance of hemodialysis arteriovenous grafts to prevent thrombosis".)
Most AV grafts requiring treatment have only one stenotic site, but up to 30 percent can have two or more stenotic sites. In one study, more than 60 percent of stenotic lesions were located at the venous anastomosis and approximately 20 percent in draining peripheral veins [10]. The remainder of stenotic lesions occur within the body of the graft (ie, intragraft; 11 percent), in the central veins (7 to 8 percent), or at the arterial anastomosis (1.5 to 2 percent). Affected sites of associated stenotic lesions in AV grafts, in order of frequency (highest first), include [14,15] (see "Endovascular intervention for the treatment of stenosis in the arteriovenous access", section on 'Sites of stenotic lesions'):
●Venous anastomotic stenosis – Approximately two thirds of reported venous stenosis lesions occur at the anastomosis.
●Intragraft stenosis – Unlike stenotic lesions occurring at other sites, which are due to neointimal hyperplasia, areas of narrowing within the AV graft are the result of the deposition of fibrin, lipid, and cellular debris.
●Peripheral draining vein stenosis – Stenotic lesions can occur in draining veins other than venous anastomosis. In association with AV grafts, these are primarily the result of blood flow turbulence.
●Central vein stenosis – The prevalence of central vein stenosis associated with AV grafts parallels the use of central catheters or other indwelling central venous devices.
●Arterial stenosis – Stenotic lesions of the "feeding" arteries can also occur. In one review, 29 percent of AV grafts were affected, with the artery-graft anastomosis affected in 24 percent of cases [16].
Hypercoagulable evaluation — Among patients with very early (within 24 to 48 hours of surgery) or recurrent AV graft thrombosis in the absence of underlying stenosis, we initiate an evaluation for a coagulation abnormality or hypercoagulable state given that such patients are at increased risk of abnormalities [6-8,17]. As an example, antiphospholipid antibodies (aPLs) such as lupus anticoagulants (LAs) and anticardiolipin antibodies (aCLs) increase the risk of access thrombosis [6-8]. A report of 97 patients on maintenance hemodialysis noted that those with LA activity (16.5 percent) had a higher frequency of vascular access thrombosis than patients without an LA (62 versus 26 percent) [6].
In general, the assessment should include an evaluation for abnormalities in protein S, protein C, and antithrombin-3 as well as the presence of aPL. This is discussed in detail separately. (See "Evaluating adult patients with established venous thromboembolism for acquired and inherited risk factors" and "Protein C deficiency" and "Diagnosis of antiphospholipid syndrome".)
TREATMENT OF STENOSIS — Underlying stenosis of an arteriovenous (AV) graft is an important predictor of thrombosis. The treatment of clinically significant stenosis of AV grafts is important clinically because it reduces vascular morbidity (via elective versus emergency repairs), preserves future access sites, and decreases the requirement of catheter placement and the associated adverse risks of catheter use [18,19].
Stenotic lesions can be treated percutaneously with angioplasty (and stenting, when indicated) or with surgery. Some AV graft lesions may respond better with surgical revision. For patients identified with stenosis of hemodialysis AV grafts, we recommend percutaneous angioplasty, rather than surgery as the initial procedure. Lesions unsuitable for percutaneous transluminal angioplasty or that have failed angioplasty/stenting treatment can be referred for surgical revision. The use of stents following failed angioplasty is discussed in detail separately. (See "Endovascular intervention for the treatment of stenosis in the arteriovenous access".)
Definitions for the hemodialysis access time points used in describing outcomes (eg, primary patency, assisted primary patency, postintervention patency) are provided in the figure (figure 1 and table 1).
Preemptive angioplasty — Because it is safe, effective, and easily performed, we suggest preemptive angioplasty as the initial treatment of choice for the treatment of clinically significant stenosis affecting AV grafts to prevent thrombosis and/or failure of the vascular access [11,14,20-24]. Correcting the stenosis may decrease the risk of thrombosis and improve graft patency; however, randomized trials have not proven this to be the case (figure 2) [25,26]. Representative studies that illustrate this point are reviewed briefly below for AV grafts. The indications and benefits of percutaneous angioplasty are reviewed separately. (See "Endovascular intervention for the treatment of stenosis in the arteriovenous access" and "Techniques for angioplasty of the arteriovenous hemodialysis access".)
●In a small trial, 21 patients with >50 percent AV graft stenosis who had never undergone any procedure following graft placement were randomly assigned to no treatment (13 cases) or angioplasty (8 cases) [27]. In the treatment group, patency was significantly prolonged, and the thrombosis rate was significantly decreased compared with no preemptive treatment (0.10 versus 0.44 per patient-dialysis years). The angioplasty group received an average of 3.4 treatments per patient.
●Over an eight-year period in one study, an intensive graft maintenance protocol using angioplasty reduced the thrombosis rate by 70 percent [28]. During this period, there was also a nearly eightfold decrease in the access replacement rate. The average age of usable AV grafts increased from approximately two to three years.
●In another study, a graft surveillance program using angioplasty decreased the graft thrombosis rate from 48 to 17 percent over a six-year period [29]. Repeated angioplasty treatments were required. The primary graft patency rate was 23 percent, while the assisted primary patency rate at one year was 68 percent.
●The cumulative patency of 192 AV grafts was evaluated for two screening protocols with intervention performed for abnormal findings [30]. Patients either had traditional screening consisting of regular AV graft examination and monitoring of venous pressure or traditional screening plus ultrasound examinations every three months. The secondary cumulative AV access patency was significantly longer in group 1 compared with group 2 at 6, 12, and 24 months follow-up. The number of interventions per graft in groups 1 and 2 were 2.1 and 1.3, respectively, a difference that was not significant.
●In a trial of 112 hemodialysis patients, patients were randomly assigned to monthly vascular access blood flow surveillance plus standard surveillance (dynamic venous pressure and physical examination) or standard surveillance alone, with only the blood flow group referred for angiography and angioplasty for stenosis greater than 50 percent [25]. There were no differences in time to graft thrombosis or time to graft loss between the groups, suggesting that prospective monitoring plus preemptive angioplasty did not provide any benefit.
●In a trial in which monitoring was performed using monthly static venous pressure/systolic blood pressure ratios, 64 patients were randomly assigned to observation if the monthly ratio was elevated (≥0.4) and followed by angioplasty of any identified stenoses [26]. Treatment in the observation group was only performed for thrombosis of access. Graft survival was similar in both groups.
Surgical revision — Compared with angioplasty, surgery is less desirable as the initial procedure to manage stenotic lesions associated with AV grafts because although the lesion is eliminated, new lesions frequently emerge over time and, with repeated surgery, potential venous access sites are lost. As an example, even a patch angioplasty results in the loss of a small portion of vein. Additional adverse effects include postoperative pain and the occasional need for a temporary central catheter access following the operation. Nevertheless, there are certain lesions that may not respond to angioplasty or are unsuitable for angioplasty. These should be referred for surgical revision. As an example, a tight stenosis associated with a previous thrombosis tends to respond poorly to angioplasty and generally requires surgical correction. One study found that thrombosed grafts with unsuccessful percutaneous thrombectomy almost always (92 percent of cases) have a stenotic lesion at a location other than the venous anastomosis [31]. In addition, following an unsuccessful percutaneous graft thrombectomy, successful salvage by surgical revision was feasible in only 8 percent of cases [31]. (See "Endovascular intervention for the treatment of stenosis in the arteriovenous access", section on 'Sites of stenotic lesions'.)
TREATMENT OF THROMBOSIS — There is no single best treatment for a thrombosed arteriovenous (AV) graft. Options include percutaneous or surgical thrombectomy. The approach chosen is largely based on local practice and expertise [12]. At the authors' institution, thrombosed AV grafts are most commonly treated percutaneously with thrombectomy in conjunction with balloon angioplasty of any identified underlying stenotic lesions. However, surgical thrombectomy and revision are used for grafts that thrombose within one month of their creation, or in cases for which the percutaneous approach is unsuccessful (ie, stenosis refractory to angioplasty, rapid recurrence of thrombosis).
Percutaneous thrombolysis — With a percutaneous approach, either pharmacologic agents, mechanical clot disruption, or a combination (pharmacomechanical thrombolysis) may be used, provided there are no contraindications to thrombolytic therapy (such as recent surgery, a bleeding disorder, a recent bleeding episode, or severe hypertension). If flow in the graft is restored, a fistulogram can then be performed to identify any associated stenoses, which can be treated with angioplasty [17,32-37]. If a percutaneous approach is not successful or deemed inappropriate, surgical thrombectomy and revision may be necessary. (See 'Surgical thrombectomy' below.)
The initial experience using thrombolytic agents, such as urokinase and streptokinase, yielded disappointing results [38]. However, subsequent dosing adjustments and technical advances have improved the success rate and reduced the incidence of bleeding [39,40]. As an example, use of the pulse-spray technique, which combines thrombolytic therapy with mechanical clot disruption, rapidly established access patency in over 90 percent of cases with minimal complications [39]. Fifty percent of these fistulas remained patent at one year. Although not available in all centers, the most common thrombolytic agent used is alteplase [41,42]. Alteplase is delivered into the entire portion of the thrombosed access using a 4 Fr (French) catheter. The operator waits at least 15 to 30 minutes before proceeding to clot removal and subsequent angioplasty of the stenotic lesion. Care must be taken since the thrombolytic agent can be inadvertently delivered directly into the systemic circulation [43,44].
Mechanical thrombolysis using specialized devices has been reported without use of any lytic agent. As examples, one study found that the Amplatz thrombectomy device and pulse-spray pharmacomechanical thrombolysis were similarly successful in recanalizing thrombosed grafts, with comparable primary patency rates [45]. Another study found similar technical success for mechanical clot disruption alone compared with pharmacomechanical clot disruption using urokinase, but AV grafts treated with mechanical thrombectomy had improved long-term patency [46].
A concern with percutaneous thrombolysis is the potential for clinically significant pulmonary emboli. In a review of over 650 cases, only one clinically apparent pulmonary embolus was documented (the symptoms of which resolved within 24 hours), and two patients developed transient chest pain of undetermined etiology [46]. A second well-designed study compared ventilation-perfusion scintigraphy before and after thrombolysis [47]. In this study, 13 patients had clotted AV grafts, 9 of whom underwent mechanical thrombolysis alone, and the remainder had pharmacomechanical thrombolysis with urokinase. No patient had radiographic evidence of pulmonary embolism. To reduce the possibility of large pulmonary emboli, a device has been developed that consists of a high-speed rotating cam tip that pulverizes the clot into tiny particles.
Surgical thrombectomy — Surgical thrombectomy is a relatively quick procedure, has a very low complication rate, and is initially successful in 90 percent of cases. However, failure to correct the underlying outflow stenosis, which can be accomplished using percutaneous or open techniques, will lead to rapid rethrombosis.
Although not adequately studied, it appears that surgical treatment for AV graft thrombosis is comparable to percutaneous intervention. This was shown in a meta-analysis of eight randomized studies, in which similar results with both techniques were reported for studies published after 2002 [48]. The primary patency of AV grafts after surgical thrombectomy and angioplasty ranged from 30 to 60 percent at three months and 10 to 40 percent at six months [49-51].
Surgical thrombectomy involves making a small incision overlying, typically, the venous outflow limb of the AV graft and then using a Fogarty embolectomy catheter to sequentially remove clot first from the graft and then from the venous and arterial anastomoses. A graft study (graftogram, on-table duplex ultrasound) can be performed to determine if any significant inflow or outflow stenoses are present. These can be managed using open surgical revision (patch angioplasty, jump graft). With the availability of intraoperative fluoroscopy imaging (mobile C-arm, hybrid operating room), balloon angioplasty of any identified lesions can be performed just as would be performed in an interventional radiology suite. Alternatively, in the absence of intraoperative imaging, vascular dilators can be passed through at least the venous anastomosis, the most likely site of stenosis. Interrogation of the arterial anastomosis may require extension of the incision or a separate incision and arterial limb graftotomy.
Antithrombotic therapy — Following treatment of AV graft thrombosis, recurrent thrombosis is common and, although mechanical treatment of the underlying lesion using angioplasty reduces it, does not eliminate the lesion. Medical therapies are primarily aimed at lessening neointimal hyperplasia or treating hypercoagulability.
Although entirely opinion based, we consider administering warfarin to selected patients with a hypercoagulable state. Specific management issues are discussed separately. (See "Overview of hemodialysis arteriovenous graft maintenance and thrombosis prevention".)
OUTCOMES — Reasonable outcome goals following treatment for graft thrombosis are as follows:
●A success rate, defined by the ability to use the graft at least once postprocedure, of 85 percent.
●Primary patency of 40 percent at three months after percutaneous thrombectomy.
●Primary patency of 50 and 40 percent at 6 and 12 months after surgical thrombectomy, respectively.
Terminology used to describe postintervention patency rates following percutaneous intervention for hemodialysis arteriovenous access in accordance with the Society of Interventional Radiology (SIR) reporting standards is given in the table (table 1) [52].
The overall patency following treatment of arteriovenous (AV) graft stenosis or AV graft thrombosis is briefly reviewed below [53-56].
Overall patency following treatment of stenosis — Preemptive angioplasty of the stenotic lesion in AV grafts is not a permanent treatment. Even after an excellent technical success (visual inspection at time of the angioplasty), which is close to 100 percent, the low rates of access patency for AV grafts at 3, 6, and 12 months are discouraging [10,25,49,57-60]. The patency rates for specific anatomic lesions are discussed in detail separately. (See "Endovascular intervention for the treatment of stenosis in the arteriovenous access".)
Overall patency following treatment of thrombosis — The overall primary patency after percutaneous thrombectomy is slightly better for AV grafts compared with AV fistulas (at six months, it is 25 to 40 percent for grafts compared with 20 to 40 percent for fistulas) [49,61]. The primary patency of the underlying stenotic lesion following thrombectomy is much worse. In a number of series, primary graft patency after thrombectomy and angioplasty was 37 to 63 percent at three months and 11 to 39 percent after six months [39,46,59,62-67]. A large observational study comparing more than 300 graft thrombectomies with more than 300 elective angioplasties performed at one large medical center observed a primary patency at six months of 19 percent after thrombectomy versus 51 percent after elective angioplasty [59].
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Dialysis" and "Society guideline links: Hemodialysis vascular access".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Beyond the Basics topics (see "Patient education: Hemodialysis (Beyond the Basics)")
SUMMARY AND RECOMMENDATIONS
●Arteriovenous graft dysfunction – Arteriovenous (AV) graft dysfunction and failure is common. Immediate failure in the postoperative period is usually due to technical issues. AV graft dysfunction and failure beyond the postoperative period (ie, delayed graft failure) is predominantly related to stenotic vascular lesions. AV graft failure can also be related to complications such as infection, pseudoaneurysm, or other conditions that lead to sacrifice of the graft. (See 'AV graft dysfunction and failure' above.)
●Clinical presentation – For patients with an abnormal clinical examination or abnormalities on monitoring and surveillance, the cause of the problem may be obvious on physical examination of the AV graft. Most cases of AV graft thrombosis are preceded by progressive stenosis at the vein-graft anastomosis that can be documented by access surveillance or clinical monitoring. However, up to 25 percent of AV grafts clot fairly abruptly without prior indication of critical stenosis. (See 'Clinical presentation and evaluation' above.)
●Evaluation of stenosis – Although most AV grafts requiring treatment have only one stenotic site, up to 30 percent can have two or more stenotic sites. A clinically significant stenosis in an AV graft is defined as a greater than 50 percent narrowing of the diameter with abnormal clinical findings, such as decreasing intragraft blood flow (less than 600 mL/min) or elevated static pressure within the graft. Affected sites of associated stenotic lesions in AV grafts are listed above and described in more detail separately. (See 'Evaluation of stenotic lesions' above.)
●Treatment of stenosis – Stenotic vascular lesions can be treated percutaneously with angioplasty, or with surgery. For patients identified with stenosis of hemodialysis access grafts, we recommend preemptive percutaneous angioplasty rather than surgery as the initial procedure (Grade 1B). Lesions unsuitable for percutaneous transluminal angioplasty can be referred for surgical revision. Even with excellent technical success rates, which approach 100 percent, the low rates of AV graft patency are discouraging. (See 'Treatment of stenosis' above.)
●Treatment of thrombosis – Once thrombosis of an AV graft has occurred, treatment options include percutaneous or surgical thrombectomy, in conjunction with angioplasty (balloon, patch) of any identified underlying stenotic lesions. The approach chosen is largely based upon local practice and expertise. Each institution should determine which initial approach is preferred. Following percutaneous thrombectomy, patency for AV grafts is slightly better compared with AV fistulas. Primary patency of the underlying stenotic lesion is much worse following thrombectomy compared with preemptive angioplasty. (See 'Treatment of thrombosis' above.)
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟
