Access and outcome in chronic haemodialysis: which one takes the lead - the first, the last or the one with longest lifespan? - Reply to the letter to the Editor "Chronic haemodialysis: the access determines the outcome?" by Chia-Ter Chao

We very much appreciate the comment of Chao C.-T. on the importance of differences in haemodialysis access patterns with regard to outcome [1]. As Chao C.-T. points out, haemodialysis accessrelated complications are an important cause of morbidity and mortality in end stage renal disease patients and are strongly influenced by the choice of haemodialysis access [2, 3]. Notably, permanent tunnelled cuffed catheters (PC) are associated with a higher frequency of access infections and higher mortality risk [4–8]. Since various studies have found striking differences in the vascular access routes used in different regions and countries [4, 9–11], we agree that this fact should be allowed for when comparing survival data. In our population we found a high rate of native arteriovenous fistulas (AVF) (85%, n = 227), which may contribute to the low rate of infection-related deaths (7% of all deaths) and the overall fair survival (one-, threeand fiveyear overall survival rates of 88%, 68% and 46%, respectively) as suggested by Chao C.-T. [1, 12]. However, there was no difference in number of AVF, arterio-venous grafts (AVG) or PC between survivors and non-survivors (table 1), and Cox regression did not detect significant survival differences between access routes (table 2). If AVF and PC alone were included in the analysis (n = 250), there was still no significant difference in survival between the two types of access (HR 0.791, p = 0.630). The missing statistical significance may of course be due to the small number of PC (9%, n = 23) and AVG (6%, n = 16) in our dialysis population. Further, it should be noted that the above-mentioned survival analysis refers to the primary vascular access. However, revisions and even creations of new vascular accesses during the course of dialysis are not uncommon. Thus further studies are needed to illuminate the impact of the last/most recent vascular access as well as the impact of the access with the longest in-use period on the outcome of patients on haemodialysis.