An age-based comparison of fistula location, patency, and maturation for elderly renal failure patients

Current Kidney Disease Outcomes Quality Initiative guidelines do not incorporate age in determining autogenous arteriovenous hemodialysis access placement, and the optimal initial configuration in elderly patients remains controversial. We compared patency, maturation, survival, and complications between several age cohorts (<65 years, 65-79 years, >80 years) to determine whether protocols should be modified to account for advanced age.

All patients at two teaching hospitals undergoing a first autogenous arteriovenous access creation in either arm between 2007 and 2013 were retrospectively analyzed from a prospectively maintained database. Kaplan-Meier survival and Cox hazards models were used to compare access patency and risk factors for failure.

There were 941 autogenous arteriovenous accesses (median follow-up, 23 months; range, 0-89 months) eligible for inclusion; 152 (15.3%) accesses were created in those >80 years, 397 (42.2%) in those 65 to 79 years, and 392 (41.8%) in those <65 years. Primary patencies in patients >80 years, 65 to 79 years, and <65 years were 40% ± 4%, 38% ± 3%, and 51% ± 3% at 12 months and 12% ± 5%, 13% ± 3%, and 27% ± 3% at 36 months (P < .001). Primary assisted patencies were 72% ± 4%, 70% ± 2%, and 78% ± 2% at 12 months and 52% ± 5%, 52% ± 3%, and 67% ± 3% at 36 months (P < .001). Secondary patencies were 72% ± 4%, 71% ± 2%, and 79% ± 2% at 12 months and 54% ± 5%, 55% ± 3%, and 72% ± 3% at 36 months (P < .001). Radiocephalic patencies were lowest among older cohorts; in those >80 years, 65 to 79 years, and <65 years, they were 65% ± 7%, 67% ± 4%, and 77% ± 3% at 12 months and 41% ± 8%, 51% ± 5%, and 68% ± 4% at 36 months (P = .019). Secondary brachiocephalic access patencies in these cohorts were 78% ± 5%, 80% ± 3%, and 82% ± 3% at 12 months and 68% ± 7%, 66% ± 5%, and 77% ± 4% at 36 months (P = .206). Both the age groups 65 to 79 years and >80 years demonstrated superior brachiocephalic vs radiocephalic secondary patencies (P = .048 and P = .015, respectively); however, no differences between configuration and secondary patency were observed within the cohort <65 years. Radiocephalic access maturation failure at 12 and 24 months was 25% ± 3% and 29% ± 4% in those <65 years, 32% ± 3% and 39% ± 4% in those 65 to 79 years, and 40% ± 7% and 48% ± 8% in those >80 years (P = .006). Brachiocephalic access maturation failures were 17% ± 3% and 20% ± 3% at 12 and 24 months in those <65 years, 21% ± 3% and 25% ± 4% in those 65 to 79 years, and 18% ± 5% and 21% ± 5% in those >80 years (P = .740). On multivariate analysis, coronary disease, female sex, previous ipsilateral or bilateral catheters, radiocephalic configuration, and age >65 years were associated with secondary patency loss.

Patients aged 65 to 79 years and >80 years had inferior primary, primary assisted, and secondary patency and maturation compared with those <65 years. When stratified by configuration, radiocephalic accesses demonstrated lower patency and maturation compared with brachiocephalic accesses for patients aged 65 to 79 years and >80 years and were an independent predictor of secondary patency loss.