Person:

Losak, Michael John

Introduction: Vascular access is a key component of hemodialysis (HD). Thus, in order to ensure optimal and successful long-term HD treatments, it is imperative to choose appropriate access tailored to individual patients. Although only 5% of patients have a functional or maturing arteriovenous grafts (AVG) at dialysis initiation, the proportion of patients using an AVG increases to 15% at one year [1]. The present work aims at assessing the impact of access characteristics on the cumulative patency of AVG. Methods: We conducted a five-year retrospective cohort study at the Icahn School of Medicine at Mount Sinai including all upper extremity AVG. We used multivariable Cox proportional hazards regression to model and assess patient/graft factors impacting primary and cumulative patency. We then used analysis of variance (ANOVA) testing for patent grafts to determine the differences in the number of re-interventions between graft types. Results: Sixty forearm (F) and 211 upper arm (U) grafts with geometries of either 6 or 8mm non-tapered (NT) or 4-7 mm tapered (T) were placed during this period. U-grafts had greater cumulative patency as compared to F-grafts [Hazard Ratio (HR): 0.55, p=0.01]. Nontapered grafts had less cumulative patency compared to tapered grafts [HR: 1.827, p=0.055]. When we divided grafts into four subgroups (U-NT, U-T, F-NT, and F-T) there were significant differences in cumulative patency between groups [p=0.031]: F-NT had the lowest patency [990.1±151.3 days]. In addition, we discovered that F-T grafts had the highest rate of reinterventions within a year for pseudoaneurysms. Conclusions: Location and geometry are important factors in determining cumulative patency and reinterventions of AVG. When planning to place an AVG, this information should be strongly considered.