Comparative effectiveness of drug-coated and uncoated balloon angioplasty in dysfunctional hemodialysis circuits among Medicare beneficiaries.

Abstract

The durability of percutaneous balloon angioplasty for arteriovenous fistula stenosis is short-lived. In randomized controlled trials, drug-coated balloons (DCBs) have shown promise in reducing reintervention compared with uncoated balloons (NDCBs). However, DCB device uptake and outcomes are poorly described. We assessed temporal trends, practice variation, and outcomes of outpatient drug-coated vs uncoated balloon angioplasty for arteriovenous fistulas in the United States. Using a 100% sample of Medicare fee-for-service data, we identified 349,521 outpatient dialysis access angioplasty procedures (DCB = 35,644; NDCB = 313,877) from 2018 to 2024 for trend analyses. For comparative analyses, we further identified patients aged 20 to 80 years who underwent intervention on an arteriovenous fistula from 2021 to 2024. Outcomes of drug-coated vs uncoated balloon angioplasty were assessed after inverse probability weighting. The primary outcome was access reintervention (a composite of angioplasty, stenting, or thrombectomy) through 12 months. Proportional DCB use increased steadily after Medicare coverage in 2018 Q1 (8.6%), reaching 16.1% in 2024 Q4 (difference, 7.5%). During 2023-2024, 30.0% of hospitals used DCBs in >30.0% of cases. For comparative analyses, a total of 64,140 procedures (DCB 12.2%, N = 7813; NDCB 87.8%, N = 56,327) were included. The mean age was 62.6 (12.4) years, 38.7% were female, and, following weighting, there was no residual imbalance in baseline characteristics. Through 12 months, there was no difference in cumulative incidence of reintervention (adjusted cumulative incidence function DCB 57.7% vs NDCB 56.9%, P = .22), but DCBs were associated with reduced restricted mean time lost to reintervention (adjusted restricted mean time lost difference: -8.0 days, 95% confidence interval: -11.1 to -4.9 days, P < .0001). DCB use has risen steadily but selectively since Medicare coverage. DCB use was associated with increased reintervention-free time through 12 months. These findings support the role of DCBs in delaying arteriovenous fistula reintervention among hemodialysis patients.