Salvage of Suboptimal or Occluded Arteriovenous Fistulas Using a 4 French System From the Radial Artery for Initial...

Introduction End-stage renal disease (ESRD) is a condition that has seen a large increase in prevalence in recent decades. Paralleling this increase in prevalence is the increasing number of patients requiring vascular access for hemodialysis. Arteriovenous fistulas (AVFs) and arteriovenous grafts (AVGs) are considered the procedures of choice for hemodialysis access. However, due to the suboptimal venous anatomy (<2 mm diameter, sclerotic vascular walls) and chronic medical conditions (diabetes mellitus) seen in many ESRD patients, successful AVF creation and maturation is not always possible using standard procedures. In this study, we performed primary balloon angioplasty (PBA) at the time of AVF creation with subsequent balloon angioplasty maturation (BAM) procedures in a group of patients with a large proportion of diabetes and suboptimal venous anatomy. The purpose of this study was to compare the assisted patency and survival rates in patients with suboptimal veins used to create AVFs to patients with standard vein AVFs and AVGs.  Methods Over a nine-year period, PBA during AVF creation was performed 682 times. Of these, 551 AVFs were matured in optimally sized veins using standard BAM procedures, and 131 AVFs were matured in suboptimal veins utilizing a modified approach. In the subset of patients with suboptimal venous anatomy, we performed the initial BAM procedure via the radial artery utilizing a 4 French system. Additionally, routine clinical surveillance was scheduled throughout the study period for all patients. Suboptimal veins included small (<2 mm diameter), sclerotic, accessory, or recanalized veins. During the study period, 69 AVGs were created and matured using standard graft-gram procedures. A Kaplan-Meier analysis of survival and assisted patency rates comparing the three groups were calculated utilizing data from a retrospective database and medical records. A hazard ratio and a log-rank test were calculated to assess statistical significance. Results The mean time of follow-up for all three groups (n=703) was 43.2 months. Among patients requiring hemodialysis access in the study, the fistula creation rate was 90.8%. Patients with suitable venous anatomy who underwent AVF creation with PBA and standard BAM procedures experienced higher primary assisted patency rates relative to the suboptimal vein AVF and AVG groups (p<0.0001). No difference was seen between the suboptimal vein AVF and AVG groups. Patient survival and the percentage of diabetics were comparable amongst all three groups. Conclusion Using our approach, we were able to achieve a high AVF creation rate amongst a group of patients with a large proportion of suboptimal veins and diabetes. Despite not performing as well as standard vein AVFs in regards to primary assisted patency, the patients with suboptimal vein AVFs experienced similar patency and survival rates as compared to patients receiving AVGs. This new approach enhances the ability to create AVFs in patients who would otherwise not be amenable to fistulas and may contribute to reduced complication risk and improved overall survival.