All arteriovenous access circuits, whether native vein or prosthetic graft, will inevitably fail due to the development of neointimal hyperplastic stenoses. These smooth muscle cell lesions obstruct blood flow and thereby decrease the efficiency of hemodialysis and increase the risk of access thrombosis. The etiology of neointimal hyperplastic stenosis is complex and multifactorial.1 Vascular stenoses develop in response to vascular injury and to endoluminal stresses applied to the blood vessel wall. Surgical creation of an arteriovenous access circuit causes acute traumatic injury at vascular anastomoses, and the high rate of blood flow (> 800 mL/min) through the circuit induces chronic injury to blood vessels in hemodialysis fistulas and grafts.2 The majority of chronic hemodialysis patients will experience a problem with their arteriovenous access circuit during the first 12 months after access placement.

Balloon angioplasty remains the standard of care for the endovascular treatment of neointimal hyperplastic stenoses obstructing blood flow in hemodialysis fistulas and grafts.3 However, the true effectiveness of balloon angioplasty remains questionable. Recent prospective studies have reported postangioplasty patency rates that were less than the expected patency rates described in national guidelines.4,5 The availability of ultra-high-pressure angioplasty balloons has improved technical success, but these results may be short-lived due to delayed elastic recoil of the blood vessel wall. Autonomic contraction of smooth muscle layers within the blood vessel wall can cause restenosis within minutes or hours after a successful angioplasty procedure (Figure 1). Because of its ease of use for both the patient and physician and a low procedural complication rate, balloon angioplasty remains the standard of care for treating neointimal hyperplastic stenoses.

Bare-metal stents are used in the treatment of venous anastomotic stenoses, but patency rates are often no better than with balloon angioplasty alone.6 Bare-metal stents have proven useful for treating restenosis due to elastic recoil and for managing angioplasty-induced vascular rupture and vascular dissection.7 A metal stent can maintain full expansion of the blood vessel lumen that produces tamponade to control bleeding and optimizes blood flow to reduce turbulence through the hemodialysis fistula or graft. However, expansile forces exerted by the metal stent may exacerbate neointimal hyperplasia, and stents with an open-mesh design are porous to cellular proliferation. Progressive ingrowth of neointimal hyperplasia through the metal mesh creates in-stent restenosis that obstructs blood flow through the arteriovenous access circuit (Figure 2). In summary, a metal stent can provide an effective short-term treatment for neointimal hyperplastic stenosis and other vascular problems, but long-term patency is often no better than with balloon angioplasty alone.8

Despite decades of clinical research, an effective, long-lasting treatment for neointimal hyperplastic stenoses remains elusive. Advancements in biomaterials and stent-graft design may provide a technological solution. A stent-graft provides endoluminal support to resist muscular contraction of the blood vessel wall plus a nonporous, biocompatible barrier to prevent cellular in-growth and in-stent restenosis. Vascular stent-grafts can provide effective and durable treatment for neointimal hyperplastic stenoses causing obstruction of hemodialysis fistulas and grafts. Ideally, a stent-graft should be sufficiently flexible to conform to variable anatomy and safely cross joint spaces (Figure 3). Because of the superficial location of most arteriovenous access circuits, self-expanding stent-grafts are preferable to balloon-expandable stent-grafts when used in hemodialysis fistulas and grafts.

In 2008, the FDA approved the first stent-graft for use as a primary treatment for venous anastomotic stenosis involving a polytetrafluoroethylene (PTFE) hemodialysis graft. Continuing advancements in stent-graft materials and design have improved biocompatibility, durability, and ease of use for a variety of vascular applications. Clinical experience has demonstrated that stent-grafts work well in treating neointimal hyperplastic stenoses, delayed elastic recoil, and other vascular problems that commonly occur along arteriovenous access circuits. However, the true clinical effectiveness of using stent-grafts to treat these problems has not been sufficiently substantiated to justify broad use of these expensive medical devices. Self-expanding stent-grafts are considerably more expensive than standard angioplasty balloons, ultra-high-pressure angioplasty balloons, or bare-metal stents. The cost effectiveness of using stent-grafts for hemodialysis access applications continues to be a subject of considerable debate. The data analysis described herein of the Gore REVISE clinical study will support the assertion that use of the GORE VIABAHN Device provides significant clinical benefit while reducing the overall cost of maintaining arteriovenous access.

THE GORE REVISE CLINICAL STUDY

The Gore REVISE Clinical Study (REVISE study) is a prospective randomized comparison of balloon angioplasty versus the GORE VIABAHN Device (with CBAS Heparin Surface) (Gore & Associates) as primary treatment for venous anastomotic stenosis causing dysfunction or thrombosis of a PTFE hemodialysis graft. Unique features of the REVISE study include enrollment of patients with thrombosed hemodialysis grafts and use of the GORE VIABAHN Device across the elbow joint. Few published studies have reported success rates after using stent-grafts for these common problems.

During the study enrollment period, 293 patients with dysfunctional (56%) or thrombosed (44%) PTFE hemodialysis grafts were enrolled and followed at 31 study sites throughout the United States. The etiologies of hemodialysis graft dysfunction included a low rate of blood flow, elevated venous pressure, or prolonged hemostasis time. Fifty percent of the study patients underwent balloon angioplasty, and 50% of patients were treated with balloon angioplasty plus a GORE VIABAHN Device. The patients were followed for 2 years after the date of the treatment procedure.

The demographic distribution of the 293 subjects enrolled in the REVISE study was reflective of the United States hemodialysis patient population. The study patients were an average age of 62 years and were primarily women (52%), African American (51%), and diabetic (65%). Their average time undergoing chronic hemodialysis therapy was 3.9 years, and the average age of their PTFE hemodialysis graft was 2.1 years. The majority (62%) of study patients had undergone previous interventions at the graft’s venous anastomosis (target lesion).

The target lesion was defined as a stenosis causing > 50% luminal narrowing at the venous anastomosis, or within 30 mm of the venous anastomosis, of a dysfunctional or thrombosed PTFE hemodialysis graft. All study patients underwent thorough fistulagraphy to evaluate and measure the target lesion and to identify other vascular stenoses along the arteriovenous access circuit. Patients with a significant (> 50%) secondary stenosis could be enrolled in the REVISE study if the entirety of the second stenosis was: (1) located more than 30 mm from the target lesion, (2) < 50 mm in length, and (3) had < 30% residual stenosis after angioplasty. Per study protocol, patients with a symptomatic central venous stenosis were excluded from enrollment.

All 293 study patients were initially treated with angioplasty. Upon full inflation of the angioplasty balloon at the target lesion, the patient was randomized to undergo treatment using either balloon angioplasty alone or balloon angioplasty plus a GORE VIABAHN Device. The operating physician could treat the target lesion using any type and size of angioplasty balloon. Before undergoing randomized treatment, the two groups of patients had similar characteristics of their target lesions; the mean percentage stenosis (74% vs 73%) and the mean length (24 vs 22 mm) of the target lesions were nearly identical.

At the discretion of the operating physicians, patients randomized to treatment with balloon angioplasty alone could undergo additional angioplasty until clinical success was achieved. Larger-diameter and/or higher higher- pressure angioplasty balloons could be used when appropriate. The 148 patients randomized to treatment using balloon angioplasty alone had a mean number of 2.1 balloon inflations at the venous anastomosis of their hemodialysis grafts.

Study patients randomized to the GORE VIABAHN Device group underwent placement of an appropriate-size device as described in the instructions for use document. The device was landed at least 1 cm into the graft and 1 cm into healthy vein. The device diameter was chosen based on 5% to 20% oversizing to the graft diameter, regardless of the vein diameter.

Nearly all (98%) of the 145 patients randomized to balloon angioplasty plus a GORE VIABAHN Device had successful treatment of their venous anastomotic stenosis using only one GORE VIABAHN Device.

Patients enrolled in the REVISE study were followed for a period of 2 years or until surgical revision or abandonment of the target lesion, the venous anastomosis of the hemodialysis graft. During the 2-year follow-up period, each patient’s hemodialysis graft was managed according to local protocols. Specific management of restenosis, new stenoses, or any other problems with each patient’s hemodialysis graft was determined by local nurses and physicians. The intent of the REVISE study was to obtain information and results that are reflective of real-life vascular access care provided to the majority of hemodialysis patients.

RESULTS OF THE REVISE STUDY

The REVISE study is the first prospective study of stent-grafts for this application that included patients with thrombosed hemodialysis grafts. The inclusion of patients with thrombosed grafts is important because it is commonly believed that these patients often have worse outcomes after any treatment. Nearly one-half of the patients (44%) enrolled in the REVISE clinical study had a thrombosed PTFE hemodialysis graft.

As reported in Table 1, patients who were in the group treated with balloon angioplasty plus a GORE VIABAHN Device had statistically superior primary patency rates at the target lesion at 6 months when compared to treatment using balloon angioplasty alone (53% vs 36%; P = .008). Patients treated with a GORE VIABAHN Device also had better primary patency of the entire arteriovenous access circuit at 6 months (43% vs 29%; P = .035). At 24 months, there was no statistical difference in the secondary patency rates between the GORE VIABAHN Device group and the angioplasty alone group (69% vs 67%).

However, the 148 patients initially treated with balloon angioplasty alone needed 61 stents or stent-grafts (53 of which were GORE VIABAHN Devices) to maintain secondary patency of their hemodialysis grafts during the 2-year follow-up period. These patients also needed 43% more angioplasty and twice as many surgical revisions to maintain hemodialysis graft patency for 2 years. If the use of stents and stent-grafts was considered a loss of secondary patency, the results for the angioplasty alone group would drop from 67% to 35%. Thus, the GORE VIABAHN Device was necessary to maintain secondary patency in both treatment arms.

Results of the REVISE study substantiate the belief that graft thrombosis portends shortened patency, whether treatment was with balloon angioplasty or balloon angioplasty plus a GORE VIABAHN Device. Study patients with dysfunctional hemodialysis grafts had better target lesion primary patency at 6 months compared to patients with thrombosed grafts, whether receiving a GORE VIABAHN Device (65% vs 36%) or balloon angioplasty alone (46% vs 24%). Use of a GORE VIABAHN Device improved primary patency for both groups of patients, dysfunctional grafts and thrombosed grafts, compared to balloon angioplasty alone. Based on these superior results, the FDA approved use of the GORE VIABAHN Device for treatment of both dysfunctional and thrombotic hemodialysis grafts.

The 2-year results from the 293 patients enrolled in the REVISE study provide interesting information about this patient population. Study patients who had no previous interventions at the target lesion had similar results, whether treated with balloon angioplasty alone or balloon angioplasty plus a GORE VIABAHN Device. Study patients who had at least one previous intervention at the target lesion had better primary patency rates at 6 months after treatment with the GORE VIABAHN Device compared to treatment using only balloon angioplasty (54% vs 29%). These results suggest that the GORE VIABAHN Device is a good option if a second intervention is needed at the target lesion. However, the REVISE study also showed that early use of a GORE VIABAHN Device significantly reduced the number of additional interventions at the target lesion compared with the angioplasty alone group (2.7 vs 3.7; P = .009) and along the arteriovenous access circuit (3.7 vs 5.1; P = .053) needed to maintain 2-year secondary patency of the hemodialysis graft.

By reducing the number of reinterventions, it is estimated that use of a GORE VIABAHN Device could provide a cost reduction of approximately $2,000 per patient.

Twenty-five patients had a GORE VIABAHN Device positioned across their elbow joint. The 6-, 12-, and 24-month primary patency rates at the target lesion were 72%, 56%, and 32%, respectively. The 6-, 12-, and 24-month secondary patency rates were 95%, 95%, and 83%, respectively. During the 2-year follow-up period, there were no reported fractures of the GORE VIABAHN Device, and the long-term patency rates were superior to treatment of a venous anastomotic stenosis using balloon angioplasty alone.

SUMMARY

The REVISE study demonstrated that the use of a GORE VIABAHN Device to treat venous anastomotic stenosis provided superior patency and fewer reinterventions when compared to balloon angioplasty alone. Early results of the REVISE study suggest that early use of a GORE VIABAHN Device is a cost-effective method to maintain long-term patency of a dysfunctional or thrombosed hemodialysis graft.

Thomas Vesely, MD, is with Vascular Access Services, LLC, in Saint Louis, Missouri. He has disclosed that he is a consultant for W. L. Gore & Associates, Inc. Dr. Vesely may be reached at (314) 863-1716; tmvesely@gmail.com.

Anthony Rodriguez, PhD, is an employee of Gore & Associates in Flagstaff, Arizona.

  1. Roy-Chaudhury P, Sukhatme VP, Cheung AK. Hemodialysis vascular access dysfunction: a cellular and molecular viewpoint. J Am Soc Nephrol. 2006;17:1112-1127.
  2. Li L, Terry CM, Shiu Y-TE, Cheung AK. Neointimal hyperplasia associated with synthetic hemodialysis grafts. Kidney Int. 2008;74:1247-1261.
  3. NKF-K/DOQI Clinical Practice Guidelines for Vascular Access. m J Kid Dis. 2006;48:S234-S246.
  4. Haskal ZJ, Trerotola SO, Dolmatch B, et al. Stent graft versus balloon angioplasty for failing dialysis-access grafts. N Engl J Med. 2010;362:494-503.
  5. Vesely TM, Siegel JB. Use of the peripheral cutting balloon to treat hemodialysis related stenoses. J Vasc Intervent Radiol. 2005;16:1593-1603.
  6. Kariya S, Tanigawa N, Kojima H, et al. Peripheral stent placement in hemodialysis grafts. Cardiovasc Intervent Radiol. 2009;32:960-966.
  7. Yamamoto Y, Nakamura J, Nakayama Y, et al. Relationship between the outcomes of stent placement and the properties of arteriovenous graft outflow vein stenotic lesions. J Vasc Access. 2012;13:426-431.
  8. Yevzlin A, Asif A. Stent placement in hemodialysis access: historical lessons, the state of the art, and future directions. Clin J Am Soc Nephrol. 2009;4:996-1008.