Redo Hemodialysis Access in Elderly Patients has Acceptable Outcomes With Similar Patency of Arteriovenous Fistulas as Compared to Grafts

Objectives: Selecting optimal hemodialysis access in elderly patients remains challenging, particularly in those requiring new options after failed initial access. We sought to describe the outcomes of redo hemodialysis access in elderly patients. Methods: All patients aged ≥65 undergoing hemodialysis access placement from 2014-2019 were retrospectively identified in the electronic medical record. Characteristics and outcomes of those with initial versus redo access were compared. Patency was depicted utilizing Kaplan-Meier methods, with censoring at loss to follow-up or death, and unadjusted Cox regression. Results: Overall, 211 elderly patients undergoing 257 procedures were included in the study. Of these, 116 (45.1%) were redo access procedures. There were no demographic or comorbidity differences between the two groups with the exception of central venous stenosis which was more common in the redo cohort (27.2% vs. 6.4%, P < 0.001). 91.5% of initial, vs. 60.3% of redo, procedures were arteriovenous fistulas (P < 0.001). Distribution of fistula type differed between the two groups with first time and redo procedures of 25.5% vs. 6.9% radiocephalic, 28.4% vs. 7.8% brachiocephalic, and 35.5% vs. 37.1% brachiobasilic respectively (P < 0.001). At 12 and 24 months, 63.6% and 44.0% of first-time accesses remained patent vs. 51% and 29.0% of redo accesses (HR 1.37, 95% CI 1.05-1.80, P = 0.02). However, there was no difference in primary patency between redo grafts and fistulas (48.7% fistulas vs. 55.0% grafts at 12 months, P = 0.47). Conclusions: These results demonstrate acceptable outcomes of redo access in elderly patients. There is no evidence from this study that prosthetic grafts are preferential, suggesting elderly patients with meaningful life expectancy who require redo access should be offered autogenous options when possible.