Upper Extremity Arteriovenous Grafts are Less Likely to be Abandoned Compared to Autogenous Fistulas Despite a Higher Reintervention Rate

Background

Upper-extremity arteriovenous (AV) access often requires reintervention. However, the frequency of reinterventionsand subsequent access failure is not well-characterized. Our goal was to evaluate the frequency and type of reinterventions, risk-factors, and outcomes after AV access creation.

Methods

We performed a retrospective review of index upper extremity AV access creations (2017–2019) within the VQI Medicare-linked Vascular Implant Surveillance and Interventional Outcomes Network dataset for patients on hemodialysis (HD). Reinterventionswere defined as open or endovascular procedures on the access occurring at 1 day or more after access creation. Access abandonment was defined as any new access creation, peritoneal dialysis, kidney transplant, or death following index access creation. Univariable, multivariable, Kaplan-Meier, and Cox regression analyses were performed.

Results

There were 2,551 patients with an index AV graft (AVG) (19.5%) or AV fistula (AVF) (80.5%). Patients who underwent an AVG were more likely older, female sex, of non-White race, nonambulatory, not living at home, and to undergo the procedure as an inpatient (P < 0.05). Reintervention rates were 1.64/person-year for AVG and 1.17/person-year for AVF. On Kaplan-Meier analysis, freedom from new AV access creation at 3 years was 72% for AVG and 78% for AVF (P < 0.001). Freedom from tunneled dialysis catheter (TDC) placement at 3 years was 66% for AVG and 71% for AVF (P = 0.19). On multivariable analysis, undergoing placement of an AVG was independently associated with an increased risk of any reintervention compared to AVF (rate ratio (RR) 1.40 95% confidence interval (CI) 1.3–1.6; P = <0.001). TDC placement was increasingly associated with each subsequent reintervention but did not vary by access type. There was an elevated risk of access abandonment with subsequent reinterventions; however, long-term access abandonment was lower with an AVG compared to an AVF (RR 0.82 95% CI 0.7–0.96; P = 0.015).

Conclusion

Reinterventions to support HD access are common, and more than 60% of patients required at least one procedure within the first year of access placement. While patients with AVG require more reinterventions, they also have a lower rate of long-term access abandonment and similar rates of TDC placement compared to patients who receive an AVF.