Vascular access practice patterns in the New Zealand hemodialysis population

Abstract

Background: Native arteriovenous fistula (AVF) prevalence varies significantly among different populations and countries. Physician practice patterns may have a strong influence on access type. We assessed differences in vascular access practice patterns across all treating centers in New Zealand. Methods: Adult (age ≥ 18 years) patients on hemodialysis therapy in the year ending September 30, 2001, were studied from the Australian and New Zealand Dialysis and Transplant Association Registry. Multinomial logistic regression was used to assess factors associated with arteriovenous graft (AVG) and catheter use. Results: Of 772 patients available for analysis, 461 patients (60%) underwent dialysis using an AVF; 122 patients (16%), an AVG; and 189 patients (24%), a catheter. On multivariable analysis, female sex (odds ratio, 5.92; P < 0.001), coronary artery disease (odds ratio, 1.89; P < 0.05), body mass index greater than 30 (odds ratio, 2.55; P < 0.05), and age (odds ratio, 1.03 per year increase; P < 0.001) were associated with an increased likelihood of AVG use. Maori and Pacific Island patients were less likely to use an AVG compared with Caucasians (odds ratio, 0.47; P < 0.05). Predictors of greater likelihood of catheter use were female sex (odds ratio, 3.9; P < 0.001), late referral (odds ratio, 1.60; P < 0.05), and age (odds ratio, 1.02 per year increase; P < 0.001). Proportions of access types varied significantly across the 7 treating centers (AVFs, 32% to 86%; AVGs, 2% to 32%; catheters, 9% to 33%; P < 0.001). After adjusting for confounding factors, significant differences persisted among access types in some centers and the national average. Conclusion: Certain patient characteristics, such as age and female sex, are associated strongly with increased AVG and catheter use. However, the significant variation in risk across centers suggests more attention needs to be given to physician practice patterns to increase AVF use rates.