An estimated 2.5 million patients worldwide receive renal replacement therapy and 450,000 of those patients are from the United States. In the mid-1990s, the National Kidney Foundation began an effort to give evidence-based guidance to the teams caring for dialysis patients. The Disease Outcomes Quality Initiative (DOQI) recommendation document was published in 1997. In 2000, the scope was expanded to include chronic kidney disease even before the need for dialysis arises. This effort is now termed the Kidney Disease Outcomes Quality Initiative (NKF-KDOQI) (1). The Vascular Access section was last updated in 2006.

Long-term dialysis access is created via the surgical construction of an arteriovenous (AV) shunt. This connection between an artery and a vein may be created by using native conduits (native arteries and veins), artificial conduits (grafts), or hybrid conduits. This chapter will address native arteriovenous fistulae (AVF).

The NKF-KDOQI document recommends that when possible, the order of preference for site of AVF placement should be (a) radiocephalic (wrist), (b) brachiocephalic (elbow), and (c) transposed brachial-basilic. An arteriovenous graft (AVG) should only be considered if AVF creation is not possible.

Regular assessment of AVF should be performed to detect hemodynamically significant stenosis because it has been proven in a randomized trial that treatment of hemodynamically significant anatomic stenoses in fistulae reduces the incidence of access thrombosis and improves patency (2). The following complementary methods may be used as part of a quality assurance program. Physical examination and evaluation for such findings as arm swelling, development of collateral veins, prolonged bleeding, change in flaccidity of fistula, and change in outflow vein should be performed by a qualified health care professional on a monthly basis, which is termed monitoring. If special instrumentation is used to perform evaluation of the AVF at periodic intervals, it is termed s urveillance. Direct flow measurements and duplex ultrasound are preferred methods of surveillance. A detailed discussion of each method is beyond the scope of this chapter, and readers requiring more information should refer to the NKF-KDOQI document. It is important to note that a single abnormal test result should not trigger a response; it is an abnormal trend or persistent abnormality that should result in further steps.