Dialysis Unplugged - Fistula First: The Naysayers Were Wrong

Editor – Dr. Jay Wish has had a career-long involvement in advocating for quality care in dialysis. Jay is Professor of Medicine at Case Western Reserve University and Medical Director of the Dialysis Program at University Hospitals Case Medical Center in Cleveland, OH.  He is past-president of the Forum of ESRD Networks and has served on several technical expert panels for CMS regarding quality oversight of the ESRD Program, including chairman of the ESRD Clinical Performance Measure-Quality Improvement Committee from its inception in 1998 to 2006. In 1997, Dr. Wish also served on the workgroup for the National Kidney Foundation for developing original evidence-based clinical practice guidelines.  In recent times, Jay served on the steering committee to develop ESRD clinical performance measures for the National Quality Forum under a contract from CMS. He is currently the nephrology consultant to the Fistula First Breakthrough Initiative. Dr. Wish is also a member of the Board of Directors of the American Association of Kidney Patients and was recipient of the 2005 Visionary Award.  He is Vice Chairman of the Editorial Advisory Board of Nephrology News & Issues and has had many articles, abstracts, reviews, and book chapters published, especially in the areas of ESRD quality oversight/improvement, accountability and anemia management.

COMMENTARY – Dr. Jay Wish

In 2003, recognizing the abysmal prevalent 33% arteriovenous fistula (AVF) rate among hemodialysis patients in the US, the superiority of AVFs vs. alternate forms of vascular access with regards to patient outcomes and costs, and the fact that other industrialized countries were achieving prevalent AVF rates in excess of 90%, CMS began the National Vascular Access Improvement Initiative with the KDOQI goals of 50% prevalent AVF rate and 10% prevalent long-term catheter rate.

In 2005, CMS renamed the program the Fistula First Breakthrough Initiative (FFBI) with a “stretch” goal of 66% prevalent AVF rate. The FFBI is organized as a coalition of stakeholders including CMS, ESRD Networks, dialysis providers, and relevant organizations including nephrologists, surgeons, interventionalists, patients, and scientists. At first the 66% prevalent AVF goal was met with considerable skepticism as being unrealistic, non-evidence-based, and another intrusion by government on clinical practice. Yet, as of Sept. 2011, the prevalent AVF rate in the US has risen to 59.7%, and two entire ESRD Networks have achieved the 66% prevalent AVF goal for their regions.

So why the continued skepticism by many regarding the mission of FFBI?

It has been alleged that FFBI promotes the placement of AVFs in all hemodialysis patients, a “one size fits all” mandate that is the antithesis of a patient-centered approach. Yet “Fistula First” is not the same as “Fistula Only”. A 66% prevalent AVF goal recognizes that 34% of hemodialysis patients will have other forms of vascular access, including around 10% with long-term catheters, around 5-10% with catheters and maturing permanent vascular access, and around 15-20% with arteriovenous grafts (AVGs). The current distribution (Sept. 2011) is 20.5% total catheters, 7.9% catheters >90 days, and 19.8% AVGs.

FFBI acknowledges that some patients are not suitable candidates for AVF placement due to vascular disease, medical instability, or limited prognosis, but this is unlikely to constitute more than 34%.

Successful placement and use of an AVF is often not easy and requires patience. The primary patency rate is only around 30%, meaning 70% of patients with an AVF will require additional procedures. Even with additional procedures, the secondary patency rate of AVFs is around 70%, meaning that 30% of AVFs never work. Some of these patients will end up with an AVG, and some of these patients will have an AVF placed and succeed in a more proximal vessel. However, the attempt to place an AVF that ultimately proves unsuccessful is no more a “waste” than unsuccessful interventions in other areas of medicine. The validity of the decision to place an AVF is based on the soundness of the population evidence that drove the decision, not the outcome.

In the majority of hemodialysis patients the decision to place an AVF is evidence-based, and should be pursued. The additional time spent with a catheter while the fistula matures and undergoes interventions has been used as an argument against placing AVFs in patients without ideal vessels. Many authors and speakers and national meetings have alleged that the prevalent catheter rate has risen since the introduction of FFBI. However, the data do not support this position. Between 2003 and Sept. 2011, the prevalent catheter rate (including patients with maturing but not usable AVFs and AVGs) has decreased from around 27% to 20.5%. That means the increased efficiency of referring incident hemodialysis patients for permanent vascular access has trumped the delay in achieving functional permanent vascular access as that access has shifted from around 40% AVGs in 2003 to around 60% AVFs in 2011.

So where do we go from here?

With a prevalent AVF rate just short of 60%, the US is tantalizingly close to the 66% goal and, interestingly, the rate of increase in prevalent AVFs has not slowed as some might have predicted because the “low hanging fruit” has been exhausted, and remains around 3% per year. Nephrologists are the “captain of the ship” of vascular access and have been increasingly active in leading the QAPI programs at the dialysis facility to improve AVF rates.

But there remain opportunities in the CKD arena for more timely vascular access placement. The incident catheter rate for hemodialysis patients has remained essentially unchanged in the 70-80% range since FFBI was implemented. Of note is when initial hemodialysis vascular access and duration of nephrology care are analyzed from the 2728 Medical Evidence Form, patients who have been under a nephrologist’s care for 6-12 months are more likely to initiate hemodialysis with an AVF in place than patients under a nephrologist’s care for >12 months.

A possible explanation for this is that patients who are under a nephrologist’s care for <12 months have been referred with late-stage CKD for preparation for renal replacement therapy, and vascular access referral is part of the care path.

On the other hand, patients who are under a nephrologist’s care for >12 months were probably referred when they had earlier-stage CKD and preservation of renal function was more of a priority. In such cases, the nephrologist may be unrealistic regarding the success of the renal preservation efforts, and may be inclined to continue those efforts to the exclusion of vascular access referral despite increasing evidence that renal replacement therapy is becoming inevitable.

There has been concern that bundling of dialysis payment would lead to an increase in AVGs as providers become more impatient that maturing AVFs and the accompanying extended catheter use increases costs from antibiotics, thrombolytics, and hospitalization-induced absenteeism. Fortunately, that has not proved to be the case in 2011 since bundling began. Nephrologists must remain vigilant to assure that vascular access choice is made on the basis of patient-centeredness and not facility expediency.