Tapered arteriovenous grafts do not provide significant advantage over nontapered grafts in upper extremity dialysis access

Objective: Although tapered dialysis access grafts are often used in an effort to prevent ischemic steal, their efficacy is uncertain. Our goal was to use real-world data to assess the performance of these grafts with respect to primary patency and ischemic steal. Methods: The Vascular Quality Initiative database was queried from 2010 to 2017 for all patients undergoing tapered dialysis grafts in the upper arm. Multivariable analysis was performed to analyze primary patency, ischemic steal, and reinterventions. Results: We identified 3608 patients who received dialysis access grafts, 1473 tapered grafts and 2135 nontapered grafts. The mean age was 64.8 years, and 43.4% of the patients were male. Tapered grafts were used more often in female patients (60.5% vs 54%), nonwhite patients (53.3% vs 47.7%), patients with no previous access (28% vs 26.3%), grafts with an antecubital brachial artery origin (50% vs 44.4%), and grafts with an antecubital cephalic vein target (7.4% vs 3.7%; P <.05). Three-month outcomes between tapered and nontapered grafts were similar for wound infection (1.4% vs 2%; P =.31), ischemic steal (4.1% vs 4.6%; P =.58), and arm swelling (3.5% vs 2.9%; P =.38). Multivariable analyses revealed that in comparison to nontapered grafts, tapered grafts did not affect primary patency rates (hazard ratio [HR], 1.17; 95% confidence interval [CI], 0.96-1.42; P =.11), ischemic steal (HR, 1.03; 95% CI, 0.64-1.65; P =.92), difference in endovascular reintervention (HR, 1.08; 95% CI, 0.74-1.16; P =.5), or operative reintervention (HR, 1.25; 95% CI, 0.86-1.82; P =.24). Conclusions: Tapered grafts for upper extremity arteriovenous access do not affect primary patency, development of steal, or endovascular reintervention in comparison to nontapered grafts. Our findings do not support the routine use of these grafts in dialysis access to improve outcomes.