Reintervention Burden and Patency of Hemodialysis Vascular Access in a 15-Year Single-Center Cohort.

Abstract

Objective: Maintaining a hemodialysis access typically requires repeated intervention, yet the cumulative reintervention burden-as distinct from patency-is rarely quantified, and configuration-specific estimates from Asian centers are scarce. We evaluated patency and quantified the configuration-specific reintervention burden of upper-extremity arteriovenous accesses over 15 years at a single center.

Methods: We retrospectively analyzed 871 consecutive upper-extremity arteriovenous accesses (648 fistulas; 223 grafts) created in 810 patients between 2009 and 2024. Every subsequent procedure was adjudicated from operative records as patency-maintaining or access-abandoning; operations unrelated to the access under study, including kidney transplantation, catheter procedures, and peripheral arterial interventions, were excluded. Patency was estimated by the Kaplan-Meier method according to accepted reporting standards and compared across configurations-radiocephalic fistula, brachiocephalic fistula, basilic vein transposition, and graft. The reintervention burden was expressed as interventions per access-year. Predictors of loss of primary patency were assessed by multivariable Cox regression, and a competing-risk analysis treated kidney transplantation as a competing event.

Results: Median follow-up was 14.7 months. Overall primary, assisted-primary, and secondary patency were 60.3%, 50.5%, and 44.6%; 87.1%, 81.5%, and 77.1%; and 96.8%, 95.7%, and 94.1% at 1, 2, and 3 years, respectively. One-year primary patency differed by configuration: 70.1% for radiocephalic fistulas, 63.3% for brachiocephalic fistulas, 71.6% for basilic vein transpositions, and 36.2% for grafts (P < .001). Across 2211 access-years, 743 patency-maintaining interventions were performed (0.336 per access-year), and 37.2% of accesses required at least one reintervention. The burden was concentrated in grafts: 1.09 interventions per access-year for grafts versus 0.25 for autogenous fistulas, an approximately 4.4-fold difference. Graft configuration (hazard ratio, 2.39; 95% confidence interval, 1.86-3.06), female sex (1.37; 1.10-1.71), ischemic heart disease (1.42; 1.04-1.94), and diabetes (1.26; 1.00-1.57) independently predicted loss of primary patency. Findings persisted after excluding maturation-assisting procedures and in the competing-risk analysis (graft subdistribution hazard ratio for reintervention, 2.32).

Conclusions: High secondary patency was achieved at the cost of substantial reintervention concentrated in grafts, which required roughly 4.4 times the maintenance of autogenous fistulas. Reporting an explicit reintervention burden alongside patency supports autogenous-first, configuration-aware access planning and patient counseling.