Determinants of Arteriovenous Fistula Maturation

Vascular access is the Achilles’ heel of hemodialysis. This was true for the first patient with CKD treated repeatedly by Willem Kolff in 1943, who ran out of access sites. It remains a challenge today; creation of autogenous accesses (arteriovenous fistulas [AVFs]) is characterized by high primary failure rates.

Successful maturation is most usefully defined as the state when an AVF is used successfully and reliably with needles, and any tunneled catheter placed for initiation of dialysis has been removed. In more detail, this means that dialysis unit staff members have to be able to place needles without trying several times and that needle removal has to occur without excessive bleeding. During dialysis, the AVF has to provide sufficient flow with minimal recirculation, such that the clearance goal can be achieved.

There are (at least) six contributors to the game of maturation, and it is useful to review what we perceive to be known about their contributions.

The access surgeon is in the unique position to form something entirely unnatural out of a named vein and artery with the hope that it will last for years to decades. The available data indicate that performing more than several hundred fistula surgeries per year increases maturation rates (1). Analysis of 2016 Center for Medicare Services data suggest that approximately 0.5% of United States access surgeons create >75 AVFs per year, whereas 86% create 10 or less (2). Surgical techniques have not been studied in any randomized fashion.

The inflow artery diameter in conjunction with heart rate and BP determine the potential AVF flow volume. After AVF surgery, blood flow rates may increase ten- to 80-fold from a baseline of approximately 50 ml/min in a brachial artery due to a decrease in resistance. The change in shear stress facilitates outward remodeling such that, over time, flow volume in the dialysis access circuit increases. This continues unless there is a narrow segment that limits flow. In the vast majority of humans, the brachial artery at the elbow before any remodeling is large enough to allow for blood flow rates in excess of 1200–1500 ml/min, with mean arterial pressures >70 mm Hg. Arteries with diameters <2 mm may limit blood flow to <400 ml/min depending on BP (3).

A vein as part of an AVF experiences changes in diameter as well as wall thickness. There is the general perception that veins after AVF creation remodel depending on shear stress: too little will not change vein diameter, and too much may lead to inward remodeling, whereas the right amount leads to outward remodeling. Intra-access pressures are likely an important cofactor in this remodeling process. Veins prefer nonpulsatile flow, and AVFs that mature successfully are typically characterized by a soft pulse. Venous valves as well as sites of previous vein injury (intravenous punctures, catheter wall contact, etc.) may impede outward remodeling and can represent sites of stenosis.

Patient factors may lead to individual challenges that need solutions. Postsurgical inflammation compounds hypercoagulability, and early access failure may ensue unless systemic coagulation is maintained. The AVF may be too deep for safe needle insertion in the context of obesity or individual fat pad distribution. Multiple large preexisting side branches can prevent augmentation in some AVFs, leading to “difficult” needle insertions marked by frequent infiltrations.

The process of care leading up to AVF creation as well as follow-up monitoring and procedures after access creation may differ among centers, leading to vastly differing AVF maturation outcomes, despite similar vascular anatomy (4). Timing and availability of interventional services as well as procedural approach to stenosis and thromboses influence outcomes.

The final frontier for AVF maturation is the ability of hemodialysis unit staff to place needles. It is often not enough to “deliver” a physiologically mature AVF; the access also has to survive the reality of staff turnover, recurrent learning curves, and a variable AVF knowledge base.

With the creation of an AVF, the patient with ESKD acquires “chronic vascular access disease.” Depending on the access type, the location of the recurrent stenosis may be the cephalic arch, the basilic swing point, the forearm cephalic vein juxta-anastomotic segment, or an access specific location. Although initial cannulation may be successful at first, repeated interventions as frequent as every 2–3 months are often necessary later on. In the context of Fistula First, the prevalence of AVFs in the United States dialysis population has increased from 22% in 1995 to over 63% in 2017 (5). However, as the distribution of forearm and upper arm accesses in the Dialysis Access Consortium (DAC) Trial (6) and Hemodialysis Fistula Maturation (HFM) Trial (4) shows, practitioners have moved the site of access creation toward the upper arm, with <25% of AVFs now created in the forearm. The resulting upper arm accesses are characterized by larger vessels at the outset as well as higher flows, both factors leading to achieving “maturation” in more patients. In the long run, these higher flow accesses with intermittent outflow stenosis are characterized by larger needle insertions site aneurysms and earlier loss of skin integrity. Higher-access flows are also part of the physiology leading to pulmonary hypertension and ischemic steal. Observational studies show that forearm accesses last longer than upper arm accesses (7).

In this issue of the Clinical Journal of the American Society of Nephrology, Allon et al. (8) aim to shed light on the question if an increase in intimal thickness has an effect on fistula maturation rates. By design, only arteries >2 mm in diameter were used for access creation, and two thirds of all accesses were in the upper arm with artery diameter >4 mm. Physiologically, these arteries do not have or only very minimally have to remodel to allow blood flow rates typically seen with mature upper arm AVFs. Stratification by arterial diameter showed a nonsignificant unadjusted increased odds ratio for clinical AVF maturation failure for smaller arteries, but the number of accesses was small. Prior analysis of the effect of medial fibrosis by the same group similarly did not find an effect on maturation. Findings by Kim et al. (9) in a predominantly Asian forearm radial-cephalic AVF population suggest that, for arteries with smaller diameters (approximately 2 mm), preexisting arterial intimal thickening does affect maturation, which is concordant with experimental blood flow modeling (3). Clinical studies suggest that angioplasty of these types of small and calcified arteries can lead to successful access maturation by increasing the diameter towards 4 mm (10). The findings of Allon et al. (8), therefore, should reassure practitioners that diseased arteries can be used if the diameter approaches 4 mm. Further study of arteries smaller than 3 and 2 mm is needed to delineate the role of intimal hyperplasia and medial fibrosis for successful access creation.

In a second study from the same group, Farrington et al. (11) analyze the predictive value of ultrasound examination performed within 90 days postoperatively for unassisted maturation defined as clinically successful needle insertions within 180 days after the ultrasound without presence of a tunneled catheter and no intervention except isolated superficialization. Factors found to be associated with unassisted access maturation were location in the upper arm, absence of stenosis, higher blood flow, larger diameter, and smaller depth.

Ultrasound is a noninvasive imaging technology that can contribute greatly to the understanding of vascular anatomy in the context of dialysis access. In particular, depth measurements as well as flow volume determinations in the arterial inflow can direct interventions.

One challenge with the use of ultrasound is that veins change diameters dynamically: preoperative assessment may reveal smaller vein diameters than those observed after regional or systemic anesthesia (12). Similarly, postoperative ultrasound protocols do not take dynamic enlargement with access augmentation (occlusion of the outflow) into account. Clinical practice suggests that fistula veins showing dilation during augmentation respond more favorably to access interventions and changes in access flows than veins that remain small, although controlled data on this question are not available.

The findings of Farrington et al. (11) accurately capture what happens when practitioners choose to create upper arm accesses: there are indeed more unassisted maturations. However, this choice also invites a higher number of high flow–associated later complications, such an steal, high-output heart failure, and aneurysmic degeneration of the access and (much) later, the artery.

Quo vadis? Well over $21 million in federal funding have been expended in the DAC Trial and the HFM Trial research endeavors; unfortunately, there is no strong lead toward solving the maturation conundrum, and practitioners have migrated up the arm for better outcomes. There are islands of forearm fistula success in this sea of relative failure, quite literally visible on the AVF prevalence map of the United States. The vascular access community may alternatively turn toward centers with high volume and high success in forearm access creations to study surgical technique, processes of care leading up to and after AVF creation, and timing and extent of endovascular procedures to pinpoint commonalities that are associated with successful maturation.

Disclosures

D.M.H. is a consultant for BluegrassVascular, Proteon Therapeutics, Medtronic, and Merck. No financial support was provided for this commentary.