Longitudinal outcomes of forearm versus upper arm arteriovenous fistulas.

Background: National guidelines recommend forearm arteriovenous fistulas (AVFs) over upper arm AVFs as the initial permanent vascular access for hemodialysis if consistent with the end-stage kidney disease (ESKD) Life-Plan, but comparative outcomes are underexplored. Our objective was to assess longitudinal outcomes of forearm vs upper arm AVFs in patients with advanced kidney disease.

Methods: Using multicenter data from three prospective studies (Hemodialysis Fistula Maturation [HFM] Study, PATENCY-1 [A Study of PRT-201 Administered Immediately After Radiocephalic Arteriovenous Fistula (AVF) Creation in Patients With Chronic Kidney Disease], and PATENCY-2), we conducted a cohort study of 1516 patients who underwent upper extremity AVF creation (2014-2019). Demographic factors, comorbidities, procedural details, and 3 years of longitudinal follow-up were captured. Outcomes included primary, primary-assisted, and secondary patency at 3 years, successful AVF use, and access-related hand ischemia (ARHI) interventions. Forearm vs upper arm AVF outcomes were compared using Cox regression and logistic regression models. Subgroup analyses included outcomes stratified by site volume using model interaction terms.

Results: The study population included 1059 forearm AVFs and 457 upper arm AVFs; mean age was 56.2 ± 13.4 years and 25.2% were female. The overall primary, primary-assisted, and secondary patency rates at 3 years was 26.2% (95% confidence interval [CI], 23.6%-29.1%), 57.6% (95% CI, 54.6%-60.9%), and 66.5% (95% CI, 63.6%-69.5%), respectively, with no significant differences between forearm and upper arm AVFs. Successful AVF use at 12 months was also similar between forearm (66.1%) and upper arm AVFs (70.0%) (odds ratio, 1.02; 95% CI, 0.71-1.48; P = .91). Forearm AVFs had lower risk of ARHI interventions (hazard ratio [HR], 0.36; 95% CI, 0.18-0.71; P = .003) compared with upper arm AVFs. Subgroup analyses showed that compared with upper arm AVFs, patients who received forearm AVFs at low volume sites (≤30 access creations per year) were at greater risk for loss of primary-assisted (HR, 2.03; 95% CI, 1.21-3.41; P < .001) and secondary patency (HR, 2.53; 95% CI, 1.33-4.83; P < .001). Patients receiving forearm AVFs at low volume sites also had lower AVF use at 12 months (odds ratio, 0.52; 95% CI, 0.21-1.31; P value of interaction = .03).

Conclusions: Although forearm AVFs demonstrate similar long-term patency and usability as upper arm AVFs, they are associated with lower rates of ARHI. However, outcomes for forearm AVFs seem to have associations with institutional volume-significantly poorer results are seen at low-volume centers. System-level efforts are needed to improve outcomes for forearm AVFs, which serve as a critical lifeline for end-stage kidney disease patients.