Lifting the veil: insights into vascular access options.

In this issue of CJASN, Ong et al. (1) and Lok et al. (2) address two important issues on either end of the vascular access care spectrum. Lok et al. (2) evaluate data pertinent to choosing the optimal access for those patients initiating hemodialysis, including the autologous native arteriovenous fistula (AVF) versus the prosthetic arteriovenous graft (AVG). At the other end of the spectrum, Ong et al. (1) suggest that a thigh AVG is an acceptable long-term access in patients who have lost all upper extremity options for a permanent access. Although AVF is considered the preferred access in themajority of patients, there is growing appreciation that the costs and effort of establishing such an access are considerable and that the attempt perhaps should not be made in every patient initiating or receiving hemodialysis. Undoubtedly, the increased impetus on AVF creation is to minimize use of the “least preferred access,” that is, long-term use of tunneled cuffed catheters (TCCs). This has certainly improved the rate of fistula use in the United States; however, overzealousness may have led to the unintended consequences of higher primary fistula failure rates and, in some instances, prolonged catheter use. An important question remains regarding whether achieving one functioning fistula, even at a premium, is worth it. Thus far, the benefits of establishing an AVF have been measured in terms of long-term outcomes, patency, reduced interventions, and infections.However, the cost of obtaining this longer-term benefit should factor into the shorter-term issues of the increased number of procedures within the first 6 months to 1 year (e.g., high primary failure and failure-to-mature rates), initial increased dependency on longer catheter use, and much later onset of chronic problems related to aneurysm formation, central venous stenosis, and high output states on the heart. There is little discussion of patient “fistula fatigue” due, in part, tomultiple failed procedures to construct or mature the AVF (for many patients, it is cosmetically unattractive) as well as to procedures related to aneurysms and occurrence of life-threatening hemorrhagic episodes. Similarly, there are little data to compare morbidity associated with postprocedure vascular and neurologic deficits of vascular access options. Finally, failure to achieve “recommended” fistula rates places programs and dialysis centers in the “public spotlight,” with possible penalties for health care providers (3). The report by Lok et al. (2) adds to increased questioning of the unwavering support of the “fistula first” approach for all hemodialysis patients. This report highlights the limitations of the current recommendations based on historical data and prods us to reconsider the merits of the fistula first approach. Six years ago, one of us coauthored an article on the clinical epidemiology of AVFs and AVGs in which we reviewed the then-available data on the superiority of AVFs (4), which was before the publication of the National Institutes of Health–sponsored study on fistula maturation andusability (5). Similarly to this analysis, AVFs at that time lacked superiority in cumulative primary patency over grafts when primary failures were included. Due to a lack of good data, we still argued for AVFs over grafts (6). However, AVFs have a discernible advantage over grafts once the primary failure is excluded, at least in the forearm. This supremacy extends beyond patency alone because the rate of interventions, infections, and subsequent accesses needed is lower once the access matures. The role of surveillance and timely intervention is one aspect of vascular access care that is not included in the study (7,8). Although the current evidence does not robustly support prolongation of patency with surveillance, it is important to note that well designed prospective studies evaluating surveillancemethods and appropriate timely intervention are lacking. The time-honored physical examination may have differential sensitivity for identifying access flow or pressure problems in AVGs compared with AVFs, accounting for the large difference in thrombolytic rates. Either way, the question remains: Is the cost of this amenity worth it? It is likely worth it to the patient whose AVF matures promptly with few augmented procedures. In these functioningmaturedAVFs, froma purely economic (and humanistic) approach, Lok et al. (2) show that the angioplasty procedure rates to maintain patency are less than half of those in grafts and significantly fewer thrombolytic interventions are needed in AVFs than in grafts. However, it is not clear how to increase the percentage of functioning fistulae from the get-go. There is a need for better identification of patients who could benefit from a fistula. A failureto-mature score, developed and reported by Lok et al. (9), is an early step in identifying patients who may have a higher probability of failure-to-mature fistula and inwhom anAVGmay provide as good as or better initial or subsequent access. We agree with the following key principles enunciated byAllon and Lok (10) for choosing an access: likelihood of early access Division of Nephrology and Hypertension, Henry Ford Hospital, Detroit, Michigan; and Wayne State University School of Medicine, Detroit, Michigan