Factors affecting haemodialysis-access survival in a single centre retrospective cohort study

A retrospective review was undertaken in the prevalent haemodialysis population of an Australian Tertiary Renal Unit of patients who started haemodialysis between April 1994 and September 2000. The study aimed to assess the factors affecting the choice of long-term arteriovenous haemodialysis access, and the factors influencing access survival. Data were obtained from clinical charts and Australia and New Zealand Dialysis and Transplant (ANZDATA) Registry data. The minimum follow up was 3 months. There were 71 patients studied (66.2% male, 22.5% diabetic, mean age 56.9 years). Late referral (< 3 months) patients constituted 35.2%. Only 39% of patients were Australian born. Seventy-one patients had a total of 96 new vascular access operations, and 57 revision operations. Of the 71 first-access procedures, 56 were native fistulae and nine were synthetic grafts. Only 31% of long-term vascular access was in place 6 weeks or longer prior to the start of dialysis. Of 25 subsequent (new) access placements, only eight were fistulae and 17 were grafts. Diabetes, peripheral vascular disease and coronary artery disease all significantly predisposed to the placement of synthetic grafts. First native fistulae had superior survival to subsequent native fistulae. Subsequent fistulae had a significant rate of early loss, but were then stable. Survival of native fistulae was always superior to synthetic grafts (P = 0.0072). First synthetic graft survival was similar to the survival of subsequent grafts. Although native fistulae are already favoured for first access, these results suggest that native fistulae should more often be considered for subsequent (secondary) access.