Sentinel vascular access monitoring after endovascular intervention predicts access outcome.

BACKGROUND AND OBJECTIVES:: The vascular access pressure ratio test identifies dialysis vascular access dysfunction when three consecutive vascular access pressure ratios are >0.55. We tested whether the magnitude of the decline in vascular access pressure ratio 1-week post-intervention could alert of subsequent access failure.

DESIGN, SETTING, PARTICIPANTS, AND MEASUREMENTS:: The retrospective study included all vascular access procedures at one institution from March 2014 to June 2016. Data included demographics, comorbidities, vascular access features, %ΔVAPR = ((Pre-Post)/Pre] × 100% assessed within the first 2 weeks post-percutaneous transluminal balloon angioplasty, time-to-next procedure, and patency. The log-rank test compared the area under the curve, receiver operating curve, Kaplan-Meier arteriovenous graft and arteriovenous fistula survival curves. A multivariable Cox proportional hazard (CP) model was used to determine the association of %ΔVAPR with access patency.

RESULTS:: Analysis of 138 subjects (females 51%; Black 87%) included 64 arteriovenous fistulas with 104 angioplasties and 74 arteriovenous grafts with 134 angioplasties. The area under the receiver operating characteristic curve for fistula failure at 3 months was 0.59, with optimal screening characteristics of 33.3%, sensitivity of 56.1%, and specificity of 63.2%. Arteriovenous fistula with <33.3% decline compared to >33.3% required earlier subsequent procedure (136 vs 231 days), lower survival on Kaplan-Meier analysis (P = 0.01), and twofold greater risk of failure (P = .006). Area under the receiver operating characteristic for arteriovenous graft failure at 3 months had a sensitivity of 52.3% and specificity of 67.4%. Arteriovenous graft with a post-intervention vascular access pressure ratio decline of <28.8% also required earlier subsequent procedure (144 vs 189 days), lower survival on Kaplan-Meier (P = 0.04), and a 59% higher risk for failure. The area under the receiver operating characteristic curve for combined access failure (arteriovenous fistula + arteriovenous graft) at 3 months had an optimal cut-point value of 31.2%, a sensitivity of 54.6%, and a specificity of 63.1%. Access with a <31.2% drop had a 62% increase in the risk of failure (hazard ratio 1.62; confidence interval 1.16, 2.27; P = 0.005).

CONCLUSION:: The magnitude of post-intervention reduction in vascular access pressure ratio provides a novel predictive measure of access outcomes.