Objective: There is an urgent and compelling need to reduce the morbidity and expense of maintaining hemodialysis vascular access patency. This large, long-term, retrospective, multicenter study, which compared access patency of autogenous arteriovenous fistulas (AVF) and synthetic bridge grafts (AVG) created with conventional sutures or nonpenetrating clips, was undertaken to resolve conflicting results from previous smaller studies. Design: Patency data for 1385 vascular access anastomoses (clipped or sutured) was obtained from 17 hospitals and dialysis centers (Appendix). Five hundred eighteen AVF (242 clip, 276 suture) and 827 AVG (440 clip, 384 suture) were analyzed. Statistical comparisons were made with Kaplan-Meier survival analysis, log-rank test, two-sample t test, and X2 test. The Cox proportional hazards model was used to confirm Kaplan-Meier analysis. Results: Access patency (primary, secondary, overall, and intention to treat) was significantly improved in access anastomoses constructed with clips. In the intention-to-treat group, primary patency at 24 months was 0.54 for clipped AVF and 0.34 for sutured AVF, and was 0.36 for clipped AVG and 0.17 for sutured AVG. At 24 months, primary patency rate for AVF successfully used for dialysis was 0.67 for clips and 0.48 for sutures, and for AVG was 0.39 for clips and 0.19 for sutured constructs. Interventions necessary to maintain patency were significantly fewer in clipped anastomoses. Conclusion: Replacing conventional suture with clips significantly reduces morbidity associated with maintaining permanent hemodialysis vascular access. This beneficial effect may be due to the biologic superiority of interrupted, nonpenetrating vascular anastomoses.