Implementation Projects

Vascular Access

The Vascular Access Implementation Project was initiated in 2007. This project aimed to implement two vascular access CARI Guidelines:

  • Preparation and placement of vascular access
  • Selection of type of access


Barriers to timely arteriovenous fistula creation: a study of providers and patients
Lopez-Vargas PA, Craig JC, Gallagher MP, Walker RG, Snelling PL, Pedagogos E, Gray NA, Divi MD, Gillies AH, Suranyi MG, Thein H, McDonald SP, Russell C, Polkinghorne KR. American Journal of Kidney Diseases 2011;57(6):873-82
[PubMed abstract]

Summary of the project

Nine renal units in Australia and New Zealand took part in the study. During the baseline phase, the units worked together with the CARI Guidelines project officer to identify the barriers to early access creation and to develop strategies and procedures to improve the number of arterio-venous fistulas (AVF) created for new haemodialysis patients.

Results showed that

  • eGFR levels for referral to the surgeon and for access creation ranged from 5 – 10 ml/min; these values were low compared to the units’ perceived eGFR level of 15 – 20 ml/min;
  • 15 – 25% of patients were referred late
  • waiting times to see the surgeon and for access creation were less than 5 and 10 weeks respectively;

The units’ perceived barriers included:

  • absence of a central database for monitoring patients’ records;
  • lack of formal policies to guide referral of patients for pre-dialysis education;
  • surgical referral or access creation;
  • late referred patients;
  • long waiting times for surgical review and access creation;
  • and patient denial.

It was concluded that lack of an effective referral pathway was the main cause for delayed fistula formation and not surgical waiting time as previously thought.

A formalized pre-dialysis pathway was developed and implemented during the intervention phase. This pathway highlighted key stages at specific eGFR thresholds for pre-dialysis patient education (eGFR 25 ml/min), surgical referral (eGFR 20 ml/min) and vascular access placement (eGFR 15 ml/min).

Post-intervention results compared to pre-intervention showed that:

  • the proportion of patients receiving predialysis education was unchanged (65% vs 64%, p = 0.83);
  • eGFR at surgical review and at vascular access placement rose significantly (8.5 vs 10.6 & 8.1 vs 10.6 ml/min respectively, p < 0.001) post-intervention but remained lower than predialysis pathway thresholds
  • The proportion of patients commencing haemodialysis with an AVF post-intervention was unchanged (40 vs 41%, p = 0.86), with no change following adjustment for age, gender, late referral, centre and pre-dialysis education (OR 0.92, p = 0.58)
  • Three of the nine centres demonstrated increases in AVF use of 11, 15 & 20%, while 3 centres reduced AVF use between 3 and 9%
  • No interaction was seen between treating centre and the intervention phase (all P > 0.05)

Overall, the implementation of a predialysis pathway did not increase AVF use at first dialysis but was associated with higher eGFR at access placement. Longer and multifaceted interventions are more likely to be effective in increasing AVF usage at first dialysis.