To analyze the results of endovascular treatment of venous anastomotic stenosis (VAS) in humero-axillary arteriovenous grafts (HAG), comparing outcomes between patent and thrombosed HAG. A retrospective cohort study was made of endovascular treated patients because of a VAS in a HAG between January 2009 and December 2019. Group A: Thrombosed HAG secondary to a VAS. Group B: Patent HAG with a VAS detected during follow-up. Technical success was defined as residual stenosis after treatment <30%, and clinical success as satisfactory immediate dialysis after surgery. After ET a biannual clinical and ultrasound follow-up was performed. Survival analysis was performed for time-to-event data to assess patency. Group A: 55 patients. Group B: 22. There were no significative differences in demographic and anatomical factors between groups. Technical and clinical success were 100% in group B and 94.5% and 91% respectively in group A. Primary patency at 1, 6 and 12 months was: Group A: 81.8%, 22.4% y 15.7% respectively. Group B: 100%, 85.9%, 76,4% (p < 0.001). Secondary patency at 1, 6 and 12 months was: Group A: 85.2%, 45.8% y 31.3% respectively. Group B 100%, 95.3%, 95.2% (p < 0.001). Use of non-covered stents was associated with an increased risk of occlusion (HR 2.669 IC 95% 1.146-6.216, p = 0.010). A higher patency of EV performed on a patent HAG is expected. It is therefore advisable to develop surveillance programs that are capable to detect VAS before its occlusion.