Preoperative Vein Mapping is Not Associated With Arteriovenous Access Type Created But is Associated With Improved Patency

Abstract

Autogenous arteriovenous (AV) fistulas are preferred for hemodialysis access, with AV grafts (AVGs) reserved for when a suitable vein is unavailable. Preoperative vein mapping can be used to identify usable veins for autogenous access. We evaluated whether vein mapping is associated with access type and outcomes. We queried the Vascular Quality Initiative database for patients who had undergone initial AV access creation from 2011 to 2019. We evaluated the associations of preoperative vein ultrasound/venogram mapping with patient characteristics, access type, and 30-day and 1-year patency. Of 20,461 patients receiving an initial AV access, 92% had undergone preoperative vein mapping. The patients who had undergone vein mapping were more likely to be women (43.1% vs 39.8%; P = .01), outpatients (84.1% vs 81.2%; P = .002), obese (43.3% vs 39.2%; P = .002), and diabetic (62.6% vs 59.8%; P = .03). They were also more likely to have concomitant coronary artery disease (21.8% vs 18.9%; P = .008), congestive heart failure (31.3% vs 25.9%; P < .001), and concurrent tunneled dialysis catheters (43.7% vs 35%; P < .001) and to already require dialysis (55.8% vs 52%; P = .004). We found no significant difference in AVG creation between the patients who had and had not undergone vein mapping (14.2% vs 13.6%; P = .5). Furthermore, no significant correlation was found between the proportions of preoperative vein mapping and AVG creation at the center and regional levels (P = .14 and P = .1; and P = .4 and P = .11, respectively). Preoperative vein mapping was associated with higher 30-day patency (94.7% vs 90%; P = .001) and 1-year patency (90% vs 82%; P < .001) for autogenous access. On multivariable analysis, concurrent tunneled dialysis catheter use, congestive heart failure, coronary artery disease, obesity, female sex, and outpatient status were independently associated with increased preoperative vein mapping use (P < .05 for all). Preoperative vein mapping was not independently associated with AVG creation likelihood (odds ratio, 0.98; 95% confidence interval [CI], 0.81-1.17; P = .8) but was independently associated with increased 30-day patency (odds ratio, 2.23; 95% CI, 1.44-3.45; P < .001) and 1-year patency (hazard ratio, 1.56; 95% CI, 1.19-2.04; P = .001). Preoperative vein mapping was more likely to be performed in patients presenting with cardiovascular comorbidities and potential technical concerns. Vein mapping did not appear to minimize AVG creation; however, it was associated with improved autogenous access patency. Although most patients received preoperative vein mapping, increasing its use might further improve access outcomes.