Abstract Background Vascular access on haemodialysis corresponds with long term outcomes. However, the effect of access on haemodialysis transfer for peritoneal dialysis (PD) patients has not been fully explored. Methods Retrospective cohort study of incident adult PD patients from the Australian and New Zealand Dialysis and Transplant (ANZDATA) Registry who transferred to haemodialysis between 2004-2022. Associations between vascular access on transfer (central venous catheter [CVC] or arterio-venous access [AVA]) and clinical outcomes (all-cause mortality, cause-specific mortality, kidney transplantation and return to PD) were compared using Cox proportional hazards analysis and competing risk models. Results Of 6,824 patients, 65% used a CVC on transfer and 35% an AVA. Variability of access type at transfer between centres was high (range 13% to 98% for CVC). AVA transfer was associated with a longer PD vintage (1.6 versus 1.2 years, P<0.001) and inadequate PD as a cause of transfer (29% versus 15%, P<0.001). All-cause mortality was lower for AVA transfer compared to a CVC (hazard ratio (HR) 0.71, 95% confidence intervals (CI) 0.66- 0.77). The risk was lowest for infection-related mortality (HR 0.59, 95%CI 0.45- 0.77) Kidney transplantation was more likely in AVA transfer compared to a CVC (HR 1.18, 95%CI 1.05- 1,33), but return to PD was less likely (HR 0.67, 95%CI 0.59- 0.71). Results remained consistent in the competing risk analysis. Conclusions Patients who transferred with an AVA, compared to a CVC, showed better survival and kidney transplantation rates, but were less likely to return to PD.
Read full abstract