Hemodialysis arteriovenous access: Detection of stenosis and response to treatment by vascular access blood flow

Abstract

Background. Hemodialysis access to the circulation is best provided by native and synthetic arteriovenous fistulae (AVF and AVG). Thromboses caused by venous outflow stenoses prevent the long-term use of AV access. This pilot study was performed to evaluate the ability of ultrasound dilution-derived access blood flows to detect AV access stenosis and to evaluate the response to treatment. Methods. This pilot study was a single-center, prospective observational intervention trial. The monitoring technique used was ultrasound dilution access blood flow measurements performed monthly and after any intervention. Screening criteria for interventions were decrements in access flow of 20% when the flow value fell under 1000 mL/min or absolute flow of <600 mL/min. The primary intervention when flow criteria were met was biplanar venography of the access with percutaneous transluminal angioplasty (PTA) of detected stenoses. Stenoses unresponsive to PTA were sent for surgical revision. Access thrombosis was considered a study ending event. Results. Baseline access flow at study entry for AVF was 919 and 1237 mL/min for AVG. Sequential measurement of AV access flow detected AV access stenosis. PTA and surgical revision significantly restored AV access flow back toward the baseline flow measurement. Failure to restore access flow by at least 20% following intervention occurred in 14% of AVF and 21% of AVG PTA attempts. Transluminal angioplasty, once successfully performed, was required at a mean of 5.8 month intervals in order to maintain AVG flow. In contrast, AVF flow was restored for a much longer period of time following angioplasty (11.4 month follow-up at the time of study end). Compared with historic controls, which used venous dialysis pressure as the primary monitoring technique, the overall (AVF-AVG) thrombosis rates improved from 25 to 16% per patient year, and AVF thrombosis rates improved from 16 to 7% per patient year. When flow was not successfully restored, thrombosis ensued. Eight of 10 thrombosis episodes were predicted based on inability to improve access flow either as a result of stenosis treatment failure or unsuccessful referral for treatment. Conclusion. Sequential measurement of AV access flow is an acceptable means of both monitoring for the development of access stenoses and assessing response to therapy. PTAs of AVF are more durable than PTAs of AV grafts.