Impact of duplex ultrasound surveillance program on patency of prosthetic arteriovenous graft for hemodialysis: A single-center experience

Background Arteriovenous prosthetic graft (AVG) is an alternative hemodialysis vascular access choice; however, its performance is limited by a high rate of thrombosis. The aim of the study was to compare the long-term secondary patency of AVG in patients undergoing a surveillance program and the long-term secondary patency of AVG in patients with clinical assessment of AVG malfunction. Methods From 2009 to 2012, all patients with AVG entered in a duplex ultrasound (DUS) surveillance program (at 3 months and then every 6 months postoperatively) to assess AVG malfunction and/or stenosis (stenosis >50% and blood flow decrease [<600 mL/min]) and eventually treated by endovascular revascularization. AVG long-term patency in the surveillance group was compared with that obtained in a historical control group in which the malfunction was clinically detected. As secondary end point, the central vein catheter (CVC) placement after AVG thrombosis was compared in the 2 groups. Results Sixty patients were included in the study, 33 (55%) in the surveillance program and 27 (45%) in the historical group. The 2 groups had similar clinical characteristics and follow-up (59, interquartile range [IQR]: 45 vs. 56 [IQR, 40 months], P = 0.32). Fifteen (45%) AVG malfunctions were detected in the surveillance group and successfully treated (10 [66.6%] angioplasty and 5 [33.4%] angioplasty stenting). No malfunction was detected in the historical control group. By Kaplan-Meier analysis, the 5-year secondary patency was significantly higher in the surveillance group compared with the historical group: 42 ± 13% vs. 9 ± 7%, P = 0.03. By Cox analysis, the DUS surveillance was a significantly protective factor for AVG thrombosis, otherwise the use of CVC before the AVG and diabetes mellitus were AVG thrombosis risk factors. The CVC placement was significantly lower in the surveillance group compared with the historical group (14.0% vs. 42.2%, P = 0.02). Conclusions The DUS surveillance allows a greater secondary patency compared with a clinical evaluation and reduces CVC placement rate.