Comparing post-maturation outcomes of arteriovenous grafts and fistulae

Abstract

Background: Though arteriovenous grafts (AVG) mature more reliably than arteriovenous fistulae (AVF) and require fewer maturation procedures (MP) to obtain functional patency, AVG are thought to have worse function after maturation. We explored differences in post-maturation outcomes between the following groups: AVF patients who did (AS-AVF) and did not (unAS-AVF) require assisted maturation and AVG patients who did (AS-AVG) and did not (unAS-AVG) require assisted maturation. Methods: Using the US Renal Data System (2012–2017), we retrospectively identified patients who initiated dialysis with a central venous catheter, underwent AVF or AVG placement and achieved successful two-needle cannulation. Primary patency and access abandonment after maturation were compared across groups using competing risks regression methods, generating sub-hazards ratios (sHR). Results: We identified 42,664 AVF and 12,335 AVG that met inclusion criteria. A larger proportion of AVFs required interventions: 18,408 AVF (43.2%) versus 2594 AVG (21.0%; p < 0.01). Both AS-AVG and AS-AVF patients experienced patency loss at 1 year more frequently compared with unAS-AVG (67.5% & 57.5% vs 55.2% respectively). Patency loss was lowest in unAS-AVF (38.9%). These trends were robust on adjusted analysis (unAS-AVG reference, AS-AVG sHR = 1.44, p < 0.01; AS-AVF sHR = 1.08, p < 0.01, unAS-AVF sHR = 0.67, p < 0.01). AS-AVGs were more likely to be abandoned than unAS-AVGs (11.7% unAS-AVG vs 17.2% AS-AVG). Fistulae, assisted or not, had lower unadjusted rates of 1-year abandonment than grafts (8.9% AS-AVF vs 7.3% unAS-AVF). On adjusted analysis, AVF usage was protective against abandonment (unAS-AVG, reference; AS-AVF sHR = 0.67, p < 0.01; unAS-AVF sHR = 0.59, p < 0.01) while AS-AVG was not (AS-AVG sHR = 1.32, p < 0.01). Conclusions: unAS-AVF have the best long-term outcomes. AS-AVF lose primary patency at a higher rate than unAS-AVG. AVGs may be a better choice than AVFs if veins are marginal and likely to require assisted maturation. Further research is needed to identify anatomic and physiologic factors that affect long-term performance and influence conduit choice.

Quality assurance in surgical trials of arteriovenous grafts for haemodialysis: protocol for a systematic review

“…There has been a 20-year debate about the role of AVG compared with AVF, based on six RCT directly comparing two modalities 22–27. This topic has received increased attention due to both increased recognition of the success of alternatives, and the costs of failed vascular access procedures 28. Additionally, innovations in AVG design now allow cannulation within a few hours rather than waiting for 2 weeks, and thus remove the need for a central-venous catheter in patients who require vascular access for haemodialysis to be commenced or continued 22.…”

“…[22][23][24][25][26][27] This topic has received increased attention due to both increased recognition of the success of alternatives, and the costs of failed vascular access procedures. 28 Additionally, innovations in AVG design now allow cannulation within a few hours rather than waiting for 2 weeks, and thus remove the need for a central-venous catheter in patients who require vascular access for haemodialysis to be commenced or continued. 22 Determining the role of 'early-cannulation' AVG is the subject of current scrutiny, particularly in the elderly or where AVF maturation is uncertain.…”