Prediction of arteriovenous access steal syndrome utilizing digital pressure measurements

Abstract

Steal syndrome is a well-known complication of arteriovenous (AV) access placement. To assess the derangement in hemodynamics of the upper extremity after AV access creation, brachial and digital pressures were performed before and after operation. Thirty-five patients (ages 20-88 years) with end-stage renal disease requiring new upper extremity hemodialysis AV access were prospectively evaluated. Values were obtained preoperatively, on the day of surgery, and 1 month postoperatively. Follow-up at 1 year was obtained on all patients. Of the 35 patients, 19 (54%) were diabetic and 9 (26%) had had a prior AV access. The AV accesses created included the following: autogenous brachial-cephalic (n = 14, 40%), autogenous radialcephalic (n = 10, 29%), brachial-basilic transposition (n = 5, 14%), prosthetic brachial-antecubital forearm loop (n = 3, 9%), autogenous brachial-axillary saphenous vein translocation (n = 2, 6%), and 1 (3%) prosthetic brachial-axillary. After AV access creation the digital brachial index (DBI) dropped in 28 (80%) of the 35 patients. Six patients (17%) developed a symptomatic steal, 3 of which (9%) eventually required revision. In those patients without ischemic steal symptoms (n = 29) the mean DBI decreased from 0.9 to 0.7 (p < 0.01) immediately and decreased no further at 1 month. For those with a symptomatic steal the DBI decreased from 0.8 to 0.4 (p < 0.01) immediately and decreased no further at 1 month. Utilizing a DBI less than 0.6, the sensitivity was 100%, the specificity 76%, the positive predictive value 46%, and the negative predictive value 100%. Hemodynamic steal after AV access creation is very common, with symptomatic steal occurring nearly a fifth of the time. Utilizing digital pressure measurements, a DBI less than 0.6 obtained on the day of surgery can reasonably predict which patients are at risk for the development of a symptomatic steal.