Use of a stent graft along with balloon angioplasty led to greater patency at six months than did balloon angioplasty alone for the percutaneous revision of venous anastomotic stenosis in patients with a prosthetic hemodialysis graft. This was the conclusion of a randomized clinical trial published in the New England Journal of Medicine (2010;362:494-503).
“It's estimated that there are 120,000 dialysis-graft interventions to maintain graft function in the US every year,” said Ziv Haskal, MD, Chief of Vascular and Interventional Radiology at the University of Maryland, in a phone interview.
“The gold standard for treating this recurring problem is balloon angioplasty. A lot of other things have been looked at, and, to this point, none of them have shown any benefit over balloon angioplasty.” Other therapies that have been attempted include balloons with a cutting mechanism and open-mesh stents.
It's gratifying to see a randomized controlled trial on stents, as currently there is a lot of junk science on stents in the literature, said Gerald Beathard, MD, PhD, who was not involved in the study, when asked to comment in a phone interview. Dr. Beathard is Vice President of Provider Development at Lifeline Vascular Access.
“I think it's an excellent study done by some very thoughtful and prudent investigators.”
He does, however, have concerns with several points of the study. For one, there was a significant difference in procedural success between the stent-graft and the balloon-angioplasty groups (94% vs. 73%), suggesting the two groups were not really comparable, he said.
Secondly, neither group met the National Kidney Foundation (NKF) Kidney Disease Outcomes Quality Initiative (KDOQI) standard of at least 50% primary patency of the access circuit at six months, with the stent-graft group at 38% and the balloon-angioplasty group at 20%.
Dr. Beathard also questioned the emphasis on the binary restenosis rate. While this is a good endpoint, this again deviates from the KDOQI guidelines, which stress pathophysiology and not anatomy, he said. “The primary patency of the dialysis circuit is a more meaningful endpoint, clinically, than [binary] restenosis.”
The study was sponsored by Bard Peripheral Vascular. Dr. Haskal receives lecture fees from the company.
The prospective, multicenter, randomized controlled trial included individuals between the ages of 18 and 90 with end-stage renal disease (ESRD) who were on long-term hemodialysis using a synthetic arteriovenous (AV) access graft in the arm.
To be included, patients had to have one or more stenoses of at least 50% and no more than 7 cm in length at the graft-vein anastomosis, as revealed by angiography. Further, the entire lesion had to be within 7 cm of the anastomosis so that approximately 1 cm of the stent graft would extend into a nondiseased vein and 1–2 cm would extend into a nondiseased graft.
Patients with a second lesion in the access circuit that was 3 cm or less from the primary lesion and was either treated within 30 days of enrollment or was at least 30% were excluded from the study, as were patients with a second lesion more than 3 cm away from the primary lesion that was 30% or more. Patients with infected grafts, with blood coagulation disorder or sepsis, or with a concomitant medical condition likely to result in death within six months after the time of stent-graft implementation were also excluded.
A total of 190 patients at 13 sites were enrolled in the study, which took place between January 2001 and June 2004. Of the participants, 97 received stent grafts after balloon angioplasty and 93 received balloon angioplasty alone. Follow-up evaluations, including angiography, were performed at two and six months after the initial procedure.
The stent graft used was a self-expanding nitinol stent covered in carbon-impregnated expanded polytetrafluoroethylene (Flair Endovascular Stent Graft, Bard Peripheral Vascular).
“The study objective was to demonstrate that treatment with a stent graft is not inferior to treatment with balloon angioplasty regarding the primary endpoint,” which was primary patency of the treatment area at six months, the authors wrote.
A primary endpoint of noninferiority, or equivalence, is typical of industry-sponsored trials for medical devices, Dr. Haskal said.
The trial's secondary endpoints, which were superiority endpoints, included primary patency of the vascular access circuit at two and six months, percent stenosis of the treatment area at two and six months, freedom from subsequent intervention, and procedural success.
Patency of the treatment area was significantly greater in the stent-graft group than in the balloon-angioplasty group at six months—51% versus 23%.
Patency of the access circuit was also greater in the stent-graft group at six months—38% versus 20%. Additionally, stent-graft recipients were more likely to have freedom from subsequent interventions at six months—32% versus 16%—and were less likely to experience binary restenosis at six months—28% versus 78%.
The difference between stent grafts and balloon angioplasty for failing dialysis-access grafts goes beyond numbers, according to Dr. Haskal.
“We find that elastic recoil at the venous anastomosis is far more common than perhaps recognized, sometimes occurring as quickly as half an hour after [balloon] angioplasty. Certainly this [recoil] is a major mode of failure,” he said. “What we're doing with a stent graft is providing a scaffold to prevent elastic recoil.
“But the most important thing, which I think is somewhat under-recognized, is really how it changes the flow dynamics at that anastomosis.”
With the stent graft, researchers were able to change an end-to-side anastomosis to an end-to-end anastomosis nonsurgically.
“We find that the downstream development of the next stenosis is very, very slow,” Dr. Haskal said. “I and other investigators have patients who are several years out from stent-graft placement with uninterrupted function and patency of their graft.”
Though Haskal et al showed improved outcomes of stent grafts over balloon angioplasty, stent grafts are not a “panacea” for patients with failing dialysis-access grafts, said Robert Kerlan Jr., MD, and Jeanne LaBerge, MD, in an accompanying editorial (N Eng J Med 2010;362:550-552).
“The ultimate goal in hemodialysis management is to provide each patient with a vascular access that will last a lifetime. With widespread adoption of the Fistula First initiative, we are coming closer to realizing this goal,” they wrote.
There was a trend toward a higher rate of thrombotic occlusion in the stent-graft group—33%—than in the balloon angioplasty group—21%—and the follow-up period of the study was relatively short, they added.
“Finally, the study was powered to show noninferiority, and the sample size was relatively small (190 patients).
“Although this clinical trial was well designed, with careful assessment of venographic and clinical endpoints, it will be important to confirm the study findings in subsequent investigations.”
Secondary fistulas are also a viable alternative for those with a lesion at the vein-graft anastomosis, said Arif Asif, MD, who was not associated with the Haskal et al study, in a phone interview. Dr. Asif is Director of Interventional Nephrology and Professor of Medicine at the University of Miami.
Dr. Asif and his colleagues reported last year that an 87% patency at six months was achieved with secondary fistula (J Am Coll Surg 2009;209:100-105). “I'm not saying that every person is a candidate for secondary fistula, but those candidates must be identified, and these options explored.”
The Fistula First movement is good and desirable, but the US dialysis patient population grew an estimated 25% in the last six years, Dr. Haskal said.
“While the percent of patients dialyzed with grafts is desirably dropping in the US, the overall population of dialysis patients is still increasing, so you still have a need for grafts, and a significant one is for patients who cannot have native AV fistulae or have used up their options for native AV fistulae.”
Rather than providing definitive answers about stent grafts for failing dialysis-access grafts, this study seems to have opened new doors for future research, which is not a bad thing, according to experts interviewed for this article.
“My primary concern is that the data [Haskal et al] produced is being taken to suggest a drastic change in the way we treat patients,” Dr. Beathard said, who added that it's too early for that kind of conclusion.
As a condition for FDA approval of the stent graft, Dr. Haskal's team is required to do an additional study with longer follow-up data, Dr. Beathard said. “I'm looking forward to the results of that trial with a great deal of anticipation.”
Post-Approval Study for the Flair Endovascular Stent Graft (RENOVA, NCT00677235) is a two-year follow-up study looking at long-term efficacy and safety of the device, and it is under way, Dr. Haskal said.
It is hoped RENOVA will answer the question of what happens at two years as opposed to six months, and what secondary patency is like with a stent graft, said Prabir Roy-Chaudhury, MD, PhD, who is not affiliated with either the first or the follow-up stent-graft study, in a phone interview. Dr. Roy-Chaudhury is Associate Professor of Nephrology and Hypertension at the University of Cincinnati.
“I would hope that this study is the first step toward a larger goal of individualizing therapy for individual patients. I think future studies could address if you could coat the stent graft with a drug, or a chemical, or a gene.”
Research that could yield conclusive proof on whether stainless-steel stents, nitinol stents, or stent grafts are the best to use would be helpful not only in terms of efficacy and safety, but also cost, Dr. Beathard said.
“We're talking about devices that are expensive, and, as physicians, we need to do what we can to hold cost down.”