Comparison of Open and Endovascular Aneurysm Repair in Native Arteriovenous Fistulas.

ABSTRACT

BACKGROUND: Arteriovenous accesses develop aneurysms (FA) during their active use, resulting in pain, erosion, bleeding, and difficulty in cannulation. This study aims to evaluate the outcomes of open and endovascular management of single FA in arteriovenous fistulas (AVF). METHODS: A retrospective review of all upper extremity primary AVFs over 12 years was undertaken at a single center. Patients undergoing elective open and endovascular repair of a single FA were identified. Thirty-day outcomes, cannulation failure, line placement, re-intervention, and functional dialysis (continuous hemodialysis) for 3 consecutive months were examined. RESULTS: Three hundred and seventy nine patients presented with a single FA that met the requirements for intervention 126 (33%) underwent endovascular repair, and the remainder 253 (67%) underwent open repair. Preoperative fistulogram identified anatomically significant issues in 91% of the cases, and these were treated by balloon angioplasty 10% within the fistula tract, 44% within the outflow tract, and 47% in the central veins. In open repair, 57% underwent plication, 35% underwent resection and re-anastomosis, and the remainder (8%) underwent interposition grafting. In endovascular repair, successful placement of a stent was achieved in all cases with 1 ± 2 (mean ± standard deviation [SD]) covered stents (diameter 6 -8 mm) placed, achieving successful exclusion of the FA. The combination of early thrombosis and cannulation failures led to the greater need for a tunneled central line in endovascular repair (6.5% vs. 2.4%; endovascular versus open repair; P = 0.04). As a result, the mean time for establishing renewed access in the index AVF was significantly higher in endovascular repair (2 ± 3 vs. 2 ± 2 weeks, mean ± SD; endovascular open repair versus open repair; P = 0.001). In follow-up, there were more secondary interventions per year in the endovascular compared to open repair groups (3.1 vs. 1.4 secondary interventions per year; endovascular versus open repair; P = 0.04). Median functional dialysis durations were superior in the open repair (48 ± 6%, mean + standard error) compared to the endovascular repair at 5 years. (26 ± 7%; P = 0.03). CONCLUSIONS: Open repair results in a more rapid return to access use, lower need for a tunneled central line, lower secondary re-intervention rates, and superior functional dialysis durations compared to endovascular repair. Open FA repair should be considered for symptomatic single FA repairs before endovascular FA repair.