The Vascular Access Score: A Tool for Standardizing Arteriovenous Access Placement

Background: The Dialysis Outcome Quality Initiative (DOQI) has established widely accepted guidelines that emphasize arteriovenous (AV) fistula creation. These guidelines however provide no direction to the practitioner regarding which patient should receive an AV fistula or a prosthetic access. An objective, standardized approach for selecting the most appropriate location and type of arteriovenous (AV) access has not been developed. This results in significant inconsistency in practice patterns for AV access nationwide and limits the ability to compare outcomes of AV access procedures. We previously developed a scoring system to assist selection of access site and type (autogenous versus prosthetic) based on a series of 80 patients who underwent a primary AV access procedure in the upper extremity. The Vascular Access Score (VAS) is a preoperative assessment tool that is derived by totaling the weighted scores of six factors (venous anatomy, obesity, diabetes mellitus, age, comorbidities, timing of dialysis) that influence access outcome and site selection. A positive VAS (≥0) is indicative of circumstances favorable for autogenous access creation. A negative VAS (<0) suggests that circumstances are unfavorable for autogenous access success and that a prosthetic access should be placed. The purpose of this study is to prospectively validate the VAS as a standardization tool to guide successful AV access placement.
Methods: Between November 2006 and March 2007, 50 patients with end stage renal disease were evaluated independently by two surgeons for initial AV access placement. Each surgeon calculated a vascular access score for each patient prior to surgery. Kappa statistical analysis was used to determine the VAS agreement between physicians and the correlation between the surgery predicted by the VAS and the actual procedure performed. In a second group of 80 patients who underwent AV fistula creation based on their VAS, the functional AV fistula maturation rate was determined.
Results: The AV fistula creation rate was 56% (28 AV fistulae, 22 prosthetic accesses). The VAS comparison between physician graders resulted in "identical recommended treatment" in 96% (96/100 observations) of cases (r = 0.992). The correlation between the treatment recommended by the VAS and the actual procedure performed was 80% (Kappa= 68.2%). The functional AV fistula maturation rate for patients scored with the VAS was 80% (64/80).
Conclusions: The VAS is a reproducible objective scoring system to guide both AV access placement type and location. The outcomes achieved by the use of the VAS approach the goals established by the DOQI guidelines. Since patient selection criteria are clearly defined by the VAS, its use may result in more predictable outcomes for AV access between surgeons and institutions. The VAS could be a valuable tool for achieving the national benchmarks for AV fistula creation and utilization established by the DOQI guidelines.