The Vascular Access Controversies Masterclass at this year’s Charing Cross (CX) International Symposium (23–25 April, London, UK) was the setting for a Great Debate titled “EndoAVFs are a failed experiment” which saw four physicians—Tobias Steinke (Schön Klinik Düsseldorf, Düsseldorf, Germany), Robert Shahverdyan (Asklepios Klinik Barmbek, Hamburg, Germany), Shannon Thomas (Prince of Wales Hospital, Sydney, Australia), and Monnie Wasse (Rush University Medical Center, Chicago, USA)—present cases, either for or against the motion. The session highlighted that, not only is there clear division on this controversy amongst the presenters, but also in the wider CX audience. Even after all presentations had been given, when asked via a live poll if they agreed that endovascular arteriovenous fistulas (endoAVFs) are a failed experiment, the audience was split nearly down the middle; 54% disagreed with the motion, with 46% agreeing.
The first to present his argument, which agreed with the motion that they are a failed experiment, Steinke focussed on the number of secondary interventions that are needed for endoAVF maturation and maintenance. At the beginning of his presentation, Steinke stated that he feels the positive reputation that endoAVF has garnered is due to “two warriors” in the field; Shahverdyan and Alexandros Mallios (Hôpital Paris Saint-Joseph, Paris, France). However, he also added that, whilst these two clinicians may see positive results—technical success is quite high for both Ellipsys (Medtronic) and WavelinQ (BD) systems, up to 100% and 97% respectively, he states—this is both from single-centre and single-operator experiences. He opined that, in the “real world”, there is a much higher need for secondary interventions for endoAVFs, when compared to surgically-created fistulas. Looking at patient selection, Steinke argued that the focus on young, male patients led to favourable operating conditions, as this is a patient population that is going to have larger vein and artery diameter. He also argued that, with a larger, more representative patient population, there has been a higher percentage of reinterventions that were needed to mature and maintain endoAVFs.
Summarising his argument, Steinke stated that “the majority of patients undergoing endoAVF procedures will require one or more adjunctive procedures, that might be costly, to create maturation. Ignoring this creates resentment amongst patients and nephrologists, who did not expect further interventions to achieve maturation. And, the need for secondary procedures will be significantly driven by the ability of the dialysis centres to cannulate an appropriate AV [arteriovenous] access.”
Endovascular AVFs are part of the algorithm
The next argument to be presented, this time against the motion, came from Shahverdyan. The main aspect of his argument was that, not only are endoAVFs not a failed experiment, but they have also already become part of the treatment algorithm (at least at his clinic in Hamburg) for several years.
“EndoAVFs are part of my vascular access creation algorithm and it fits this ‘distal to proximal’ strategy,” he said, adding that “we know that we have high satisfaction, maturation, and usability rates.” Further, he argued that, whilst Steinke had shown that there was a lack of evidence supporting the wider use of endoAVF outside of specific centres and practitioners due to a lack of training or experience, the same is true for surgically created AVFs.
“EndoAVFs show comparable high maturation, cannulation and secondary patency rates,” he stated, adding the caveat that this was the case when they are created by “experienced hands”. The key point, he argued, is that “this also applies to the surgical fistulas. Why should we focus on the learning curve for an endoAVF but not for a surgical fistula? Learning curves and experience need to be considered for both [surgical and endoAVFs].”
Shahverdyan also argued that the opinions of patients should be taken into consideration when looking at the success of the endoAVF “experiment”. He stated that one thing that should be considered is the high rate of satisfaction for patients with endoAVFs. Without going into detail, Shahverdyan stated that “although patients who have surgical fistulas are usually satisfied, and patients who have bad fistulas do not care about which kind of fistula they have, mostly, the trend is they are very satisfied with their endoAVFs”.
He also argued that the flexibility granted from the use of endoAVFs show that it is better for patient outcomes. “More disciplines can create [endoAVFs], and if they’re experienced, more patients can get dialysis access early on, avoiding catheter-related complications”. As well as this, he argued that “it is not a replacement for surgical fistulas, it’s an addition, and it is an excellent addition”. Concluding his presentation, Shahverdyan added that “we hope that, with the next generation devices, the problems we have with the first-generation devices might and will be avoided.”
We can have better results with surgical AVFs
Next up to the plate, arguing alongside Steinke in support of the motion, was Thomas. He began by stating that this was not going to be an easy argument to make. “These devices aren’t experimental,” he stated, “they’re approved. People are using them, and I think one of the great things is, anyone who’s got a lab, a needle and a wire can make a percutaneous AVF.” However, he continued that endoAVFs are a failed experiment because they’re the wrong approach to the vascular access dilemma that clinicians are facing.
“There’s a tsunami of patients that we’re all facing,” Thomas said, “who are developing dialysis dependency and they need a quick access that’s easy to use, that’s resistant to infection, that doesn’t need a lot of upkeep.” He then stated that he would show the audience, “through some accepted truths”, that endoAVFs are not the solution, nor are they even part of the solution.
The first of the “truths” that he referred to were that there is a rising incidence and prevalence of kidney disease throughout the world, and that, for long-term haemodialysis patients, the matured AVF is the preferred option, as it has got a reduced risk of mortality, thrombosis, a longer usable lifespan, and “there are just many positives to that over other access types”. Whilst he acknowledged that maturation was an issue for surgical AVFs, he argued that better results could be achieved by “upskilling” with surgical AVFs rather than turning to endoAVFs and that early recognition of issues via the use of on-table ultrasound to assess flow rates and the presence of stenoses can be used as a predictor for maturation, allowing for earlier intervention and greater maturation rates.
EndoAVFs need to be in the guidelines
The final presentation of the debate, joining Shahverdyan in defence of endoAVFs, was given by Wasse, who was arguing for the inclusion of endoAVFs in clinical guidelines. Looking at this, she began by stating that it is “important to note what criteria need to be met for that to occur”. The process, she reminded the audience, follows six steps; formulate a question, select outcomes, perform a systematic review, grade this process, evaluate the evidence available, and then recommendations regarding inclusion can be made.
Wasse’s first point was that, when evaluating the data that appear to argue against the use of endoAVFs via the use of the Newcastle-Ottowa scale of assessment of 17 studies that have, to date, looked at over 2,000 patients who have received endoAVFs using one of the two US Food and Drug Administration (FDA) approved devices, the maximum score that any of them achieved was six out of nine, suggesting that the quality of data is not as good as one may hope it would be.
She highlighted that, whilst much of the currently available data are observational, single arm and retrospective, with a heterogeneous study design and limited follow-up in some cases, there is no doubt that endovascular technical success is high and achievable. As she put it, “there’s good cumulative patency and early cannulation feasibility in select patients. And that is what the data does, in fact, support.”
Similarly to Shahverdyan, Wasse also highlighted the high patient satisfaction with endoAVFs, as well as the fact that “studies to date really do support the idea that [endoAVFs] are non-inferior to surgical fistulas, and that in some ways, there’s a lower incidence of steal, as well as surgical site infection”.
Bringing her presentation to a close, Wasse added that the successful implementation of endoAVFs relies “on proper training, patient selection, and application”. She argued that we should not “do away” with endoAVFs because of issues with cannulation, but we should instead ensure that all team members are sufficientlytrained to use these accesses.
