OBJECTIVEThis study explored the long-term outcomes of arteriovenous fistulas treated with vonapanitase (recombinant human elastase) at the time of surgical creation.METHODSThis was a randomized, double-blind, placebo-controlled trial of 151 patients undergoing radiocephalic or brachiocephalic arteriovenous fistula creation who were randomized equally to placebo, vonapanitase 10 μg, or vonapanitase 30 μg. The results after 1 year of follow-up were previously reported. The current analysis occurred when the last patient treated was observed for 3 years. For the current analysis, the primary end point was primary patency; the secondary end points included secondary patency, use of the fistula for hemodialysis, and rate of procedures to restore or to maintain patency.RESULTSThere was no significant difference in the risk of primary patency loss with vonapanitase 10 μg or 30 μg vs placebo. When seven initial patency loss events related to cephalic arch and central vein balloon angioplasty were excluded, the risk of patency loss was reduced with vonapanitase overall (hazard ratio [HR], 0.63; P = .049) and 30 μg (HR, 0.51; P = .03). In patients with radiocephalic fistulas (n = 67), the risks of primary and secondary patency loss were reduced with 30 μg (HR, 0.37 [P = .02] and 0.24 [P = .046], respectively). The rate of procedures to restore or to maintain fistula patency was reduced with 30 μg vs placebo (0.23 vs 0.72 procedure days/patient/year; P = .03) and also reduced in patients with radiocephalic fistulas with 30 μg vs placebo (0.17 vs 0.85 procedure days/patient/year; P = .048).CONCLUSIONSIn this study, vonapanitase did not significantly improve primary patency in the primary analysis but did significantly improve primary patency in an analysis that excluded patency loss due to cephalic arch and central vein balloon angioplasty. In patients with radiocephalic fistulas, 30 μg significantly improved primary and secondary patency. Vonapanitase 30 μg decreased the rate of procedures to restore or to maintain patency in the analysis that included all patients and in the subset with radiocephalic fistulas.