Long-Term Outcomes of Arteriovenous Fistulas with Unassisted versus Assisted Maturation: A Retrospective National Hemodialysis Cohort Study

Abstract

Background: About half of arteriovenous fistulas (AVFs) require one or more interventions before successful dialysis use, a process called assisted maturation. Previous research suggested that AVF abandonment and interventions to maintain patency after maturation may be more frequent with assisted maturation versus unassisted maturation.

Methods: Using the US Renal Data System, we retrospectively compared patients with assisted versus unassisted AVF maturation for postmaturation AVF outcomes, including functional primary patency loss (requiring intervention after achieving AVF maturation), AVF abandonment, and frequency of interventions.

Results: We included 7301 patients ≥67 years who initiated hemodialysis from July 2010 to June 2012 with a catheter and no prior AVF; all had an AVF created within 6 months of starting hemodialysis and used for dialysis (matured) within 6 months of creation, with 2-year postmaturation follow-up. AVFs matured without prior intervention for 56% of the patients. Assisted AVF maturation with one, two, three, or four or more prematuration interventions occurred in 23%, 12%, 5%, and 4% of patients, respectively. Patients with prematuration interventions had significantly increased risk of functional primary patency loss compared with patients who had unassisted AVF maturation, and the risk increased with the number of interventions. Although the likelihood of AVF abandonment was not higher among patients with up to three prematuration interventions compared with patients with unassisted AVF maturation, it was significantly higher among those with four or more interventions.

Conclusions: For this cohort of patients undergoing assisted AVF maturation, we observed a positive association between the number of prematuration AVF interventions and the likelihood of functional primary patency loss and frequency of postmaturation interventions.

Keywords: arteriovenous access; arteriovenous fistula; vascular access.

Graphical abstract

Figure 1.

Patient cascade of the study cohort initiating hemodialysis (HD) with a CVC and a subsequent AVF placement and maturation.

Figure 2.

Adjusted ORs of AVF functional primary loss and abandonment and risk ratios (RRs) of frequency of postmaturation intervention. (A) Functional primary patency loss, (B) AVF abandonment, and (C) frequency of postmaturation intervention. Adjusted for age, sex, race, Liu comorbidity index, functional status, facility type, facility chain status, and log time of CVC dependency. LCL, lower confidence limit; UCL, upper confidence limit.

Figure 3.

Restricted cubic spline plot of the adjusted natural log OR of AVF functional primary patency loss versus the number of prematuration interventions with five knots. The curve ends the 99% percentile of the number of prematuration interventions, i.e., five. The light dotted curves present the 95% CIs. The straight green line is the reference line of OR of 1. Adjusted for age, sex, race, Liu comorbidity index, functional status, facility type, facility chain status, and log time of CVC dependency. The dose-response association was not significantly nonlinear (P=0.39).

Figure 4.

Cumulative number of interventions by number of prematuration interventions.