Vascular access survival and incidence of revisions: A comparison of prosthetic grafts, simple autogenous fistulas, and venous transposition fistulasfrom the United States Renal Data System Dialysis Morbidity and Mortality Study

Objective. The study's aim was to evaluate access patency and incidence of revisions in patients initiating hemodialysis and to determine differences in access performance by type of access among patient subgroups. Methods. The study used data from the United States Renal Data System Dialy sis Morbidity and Mortality Study Wave 2, which contained a random sample o f dialysis patients initiating dialysis in 1996 and early 1997. Failures an d revisions were evaluated among 2247 newly placed hemodialysis accesses by using Cox proportional hazards regression model and Poisson regression. Pr imary and secondary patency rates were estimated using the Kaplan-Meier met hod. Results. Fifteen hundred seventy-four prosthetic grafts, 492 simple autogen ous fistulas, and 181 venous transposition fistulas were available for eval uation. Prosthetic grafts had a 41% greater risk of primary failure compare d with simple fistulas (relative risk, 1.41; 95% CI, 1.22-1.64; P<.001) and a 91% higher incidence of revision (relative risk, 1.91; 95% CI, 1.60-2.28 ; P<.001). At 2 years, autogenous fistulas demonstrated superior primary pa tency (39.8% versus 24.6%, P<.001) and equivalent secondary patency (64.3% versus 59.5%, P=.24) compared with prosthetic grafts. When compared with si mple fistulas, vein transpositions demonstrated equivalent secondary patenc y at 2 years (61.5% versus 64.3%, P=.43) but inferior primary patency (27.7 % versus 39.8%, P=.008) and had a 32% increased incidence of revision (P=.0 4). Autogenous fistulas had superior primary patency compared with prosthet ic grafts in all patient subgroups except for patients with previously fail ed access. Vein transpositions showed the greatest benefit in terms of pate ncy and incidence of revision in women and in patients with previously fail ed access. Conclusions. The preferential placement of autogenous fistulas may increase primary patency and decrease the incidence of revisions. Vein transpositio ns had similar secondary patency compared with simple fistulas, but require d more revisions. The greatest benefit of a vein transposition fistula was seen in women and in patients with a history of access failure.