#1740 Survival of native arteriovenous fistulas in elderly patients

Abstract

Abstract Background and Aims Current guidelines recommend native arteriovenous fistulas (nAVF) as the first-choice vascular access for hemodialysis. However, the patient's age may make its indication doubtful. The aim of our study was to know the association between primary survival (PS), defined as the time to the first reparation, and secondary survival (SS), defined as the end of the working life of a nAVF, with the patient's age. Method We conducted a prospective observational study based on a historical database including 795 nAVF from incident patients, both treated in an in-hospital dialysis unit (HDU) and an out-of-hospital dialysis center (DC), made between 1998 and 2022 and followed-up until June 2023. We performed Kaplan Meier survival curves to estimate PS and SS of the nAVF for each age group (≤ 50, 50-70 and ≥ 70 years) and a multivariate logistic regression analysis of PS and SS adjusted by comorbidity (Charlson's Comorbidity Index without age (CCI)), sex, nAVF indication at end stage kidney disease (ESKD) clinics, location of the nAVF, in-hospital dialysis unit or out-of-hospital dialysis center treated patient and initial laboratory parameters (albumin, hemoglobin and phosphorus). Results The medium age of the patients included was 63.4 years; 68% were males, 34% had diabetes and 63.3% had a CCI ≥ 3 at the time de nAVF was made. 53.5% of all the nAVF were indicated at the ESKD clinics; 57.9% of them were radiocephalics, 33.4% brachiocephalics and 8.8% brachiobasilics. Almost 40% of the fistulas had an in-hospital use. In the survival study we didn't find statistically significant differences for age or sex, but both PS and SS were lower for patients with a CCI ≥ 3, and SS was significantly better for the first nAVF of the patient and for radio-cephalic nAVF, as shown in Fig. 1. These results were concordant to the univariant analysis, in which we didn't find differences for age group or sex, but both PS and SS were worse in patients with a CCI ≥ 3 (HR 1.32, p 0.037). Patients with lower albumin levels at the time the nAVF was made had a lower PS (HR 0.62, p < 0.001); and SS also had a statistically significant association with being treated at an out-of-hospital dialysis center (HR 0.69, p 0.024) and the indication of the nAVF at the ESKD clinics (HR 0.71, p 0.036). Nevertheless, in the multivariate logistic regression analysis we only found significant differences, both for PS and SS, regarding the albumin levels at the time the nAVF was made (HR 0.62, p 0.001). Conclusion The survival of a native arteriovenous fistula, both primary and secondary, does not correlate to the patient's age, sex, comorbidity, indication at the ESKD clinics or the location of the fistula.