Introduction This retrospective study investigated the factors and the effects of different venous outflows on forearm arteriovenous graft patency. Methods The venous outflow sites included basilic, cephalic, median antecubital, and deep veins. Comparisons among multiple groups were analyzed. Findings A total of 179 patients with forearm loop arteriovenous grafts met the inclusion criteria. Of these, 72 were basilic, 48 were cephalic, 44 were median antecubital, and 15 were deep. The median observation period was 19 months. The survival rate was 84.9% at 24 months and 78.2% at 48 months. Primary, secondary, and assisted primary patency rates for all arteriovenous grafts were 48.9%, 72.4%, and 68.4% at 12 months; 13.8%, 33.9%, and 23.6% at 24 months; and 0.6%, 4.6%, and 2.3% at 48 months, respectively. Differences in primary patency were statistically significant compared with those of secondary and assisted primary patency (P < 0.05). Primary patency rates for cephalic, median antecubital, basilic, and deep were 47.9%, 48.6%, 47.7%, and 40.0% at 12 months and 12.5%, 13.9%, 22.7%, and 0% at 24 months, respectively. Secondary patency rates for cephalic, median antecubital, basilic, and deep were 75.0%, 69.4%, 75.0%, and 73.3% at 12 months and 39.6%, 30.6%, 38.6%, and 13.3% at 24 months, respectively. There was no significant difference in primary thrombosis among basilic, cephalic, median antecubital and deep. There were no significant differences observed in primary or secondary patency rates among all the groups. Stenoses in the venous anastomosis and outflow vein were frequently observed in all types of arteriovenous grafts. Central venous stenosis was most commonly seen in deep (26.67%). On average, 1.9 interventions per patient were performed on the graft to maintain function. Conclusion Different venous outflow selections were not associated with long-term patency and the occurrence of thrombosis in hemodialysis forearm loop arteriovenous grafts.