Drug-Coated Balloon Angioplasty for Dysfunctional Arteriovenous Fistulas

Worldwide, approximately 850 million individuals have chronic kidney disease and nearly 4 million receive renal replacement therapy. Among those 4 million, 520,000 Americans are undergoing dialysis, and fewer than 225,000 have a functioning kidney transplant. The National Kidney Foundation Kidney Disease Outcomes Quality Initiative is working to develop an individualized, patient-driven life plan that will address disease progression among dialysis patients.

At present, preference is given to arteriovenous fistulas or grafts for most patients undergoing dialysis due to the lower risk of infection associated with those treatment methods compared with the use of a central venous hemodialysis catheter. However, according to Robert A. Lookstein, MD, MHCDL, and colleagues, the high incidence of dysfunction caused by vascular stenosis within the fistula circuit can lead to inadequate hemodialysis. The percentage of patients who undergo repeat intervention within 6 months is approximately 50%.

The current recommended treatment for dysfunctional hemodialysis fistulas is standard percutaneous transluminal angioplasty. However, that treatment yields poor long-term outcomes. Outcomes may be improved with the use of drug-coated balloons delivering the antirestenotic agent paclitaxel. Dr. Lookstein et al. conducted a prospective, global, multicenter, single-blind, 1:1 randomized clinical trial to evaluate the IN.PACT AV drug-coated balloon (Medtronic) in comparison with standard (non-drug-coated) balloon angioplasty for the treatment of new or nonstented restenotic lesions up to 100 mm in length in arteriovenous fistulas. Results were reported in the New England Journal of Medicine [2020;383(8):733-742].

The trial was conducted at 29 sites in the United States, Japan, and New Zealand. The primary effectiveness end point was target-lesion primary patency, defined as freedom from clinically driven target-lesion revascularization or access-circuit thrombosis measured during the 6 months following the index procedure. The primary safety end point was defined as serious adverse events involving the arteriovenous access circuit within 30 days following the procedure.

Eligible patients were at least 21 years of age and presented with a new or nonstented restenotic native arteriovenous dialysis fistula that had at least 50% stenosis. The trial device was a drug-coated balloon that carried a paclitaxel dose of 3.5 mg per square millimeter with a urea excipient.

A total of 330 participants underwent randomization: 170 were assigned to receive treatment with a drug-coated balloon and 160 were assigned to receive treatment with a standard balloon. Of the total cohort, 204 participants were treated in the United States, 112 in Japan, and 14 in New Zealand. Baseline characteristics were similar in the two groups; as expected, a high percentage of participants had diabetes, hypertension, or cardiovascular disease. The distribution of forearm (radiocephalic) and upper-arm (brachiocephalic and brachiobasilic) arteriovenous access lesions treated was even between the groups. Common presenting clinical symptoms of arteriovenous fistula dysfunction included decreased blood flow and elevated venous pressure. In  most participants, the target lesions were in the venous outflow, including the cephalic arch, with the lesion in 25.5% (n=84/330) located at the arteriovenous anastomosis.

The mean length of the balloons used in the index procedure was greater in the drug-coated-balloon group than in the standard-balloon group (50.0 mm vs 47.4 mm). The final mean percent diameter stenosis was similar in the two groups, as was antiplatelet therapy use after the procedure.

During the 6 months following the index procedure, 82.2% of participants in the drug-coated-balloon group had target-lesion primary patency compared with 59.5% in the standard-balloon group (risk difference, 22.8 percentage points; 95% confidence interval [CI], 12.8 to 32.8; P<.001). When the effect of missing data was examined in sensitivity analyses, conclusions were consistent with those of the primary analysis. In the drug-coated-balloon group, the percentage of participants with clinically driven target-lesion revascularization was 16.4% compared with 38.5% in the standard-balloon group (risk difference, –22.1 percentage points; 95% CI, –31.9 to –12.3).

In safety end point analyses assessing the percentage of participants with a serious adverse event involving the arteriovenous assess circuit within 30 days, the drug-coated balloon was noninferior to the standard balloon (4.2% and 4.4%, respectively; risk difference, –0.2 percentage points; 95% CI, –5.5 to 5.0; with a noninferiority margin of 75 percentage points, P=.002 for noninferiority). In sensitivity analyses, the conclusions were consistent with those of the primary analysis.

The researchers cited some limitations to the trial, including the inability to utilize a double-blind trial design because the drug-coated balloon has a different appearance than a standard balloon, and the fact that repeat intervention rates may be biased. Other limitations were only reporting short-term outcomes, and possible confounders due to between-group differences in the number of inflations and the maximum inflation pressures.

“Treatment of  dysfunctional native hemodialysis arteriovenous fistulas with a drug-coated balloon provided primary patency, including freedom from clinically driven target-lesion revascularization that was superior to that provided by standard balloon angioplasty. The drug-coated balloon was noninferior to standard balloon angioplasty with respect to safety,” the researchers said.

Takeaway Point

1. Standard recommended treatment for dysfunctional hemodialysis fistulas is percutaneous transluminal angioplasty; however, that treatment is associated with poor long-term outcomes.
2. Results of a prospective 1:1 randomized trial comparing standard balloon treatment with drug-coated balloon treatment demonstrated that target-lesion primary patency was maintained during the 6 months after the procedure more often in participants in the drug-coated-balloon group than in the standard-balloon group.
3. In analyses of the primary safety end point, drug-coated balloons were noninferior to standard balloons.