Randomised controlled trials (RCTs) are often considered the “gold standard” of evidence on which to base clinical decision-making, but, asked Theodore Saad (Sidney Kimmel Medical College, Philadelphia, USA), the first secretary and past-president of the American Society of Diagnostic and Interventional Nephrology (ASDIN), can they be “fool’s gold”? Receiving the society’s annual Gerald Beathard Award for “teaching excellence, scholarly activity, and clinical excellence” at its 19th Annual Scientific Meeting (17–19 February, Orlando, USA), his acceptance presentation turned the focus of the conference to the limitations of RCTs in vascular access, as well as the other evidence available to clinicians.
Opening by assuring the audience that he was “not here to diss RCTs”, Saad opined that it was among the proudest achievements of ASDIN that its members have participated in the organisation, execution, and publication of multiple important trials. “This is a massive accomplishment,” he stressed, before adding that, nevertheless, “RCTs are not the answer to everything.” He put it forward that personalised medicine offers a new paradigm, and quoted David S Jones and Scott Harris Podolsky (Harvard Medical School, Boston, USA) in a 2015 piece in The Lancet in saying that it “refocuses clinical attention away from the ‘typical’ patients analysed by RCTs and onto the idiosyncrasies, genetic or otherwise, of individual patients”.
Next, he compared two of his haemodialysis patients with arteriovenous grafts. First was a patient with consistently low access flow of >300 ml/min, , severe outflow vein stenosis, and frequent dialysis hypotension on no anticoagulation—yet whose graft had never thrombosed. With this patient he contrasted a second with “rapid, repeated graft thrombosis” despite access flow >800 ml/min, no significant stenosis of the access circuit, frequent dialysis hypotension, on full anticoagulation. Saad commented: “The second patient should not have been clotting while the first should have been clotting all the time—so there was something hugely different in the biology of these patients that I just did not understand.” Including either of these patients in an RCT studying an intervention such as stent-graft or drug-coated balloon angioplasty, would cause “some real muddling of the data,” he said.
Rather than a total reliance on RCT, then, there is a “montage” of evidence required in clinical decision-making, Saad argued: “We make the best decisions we can with various inputs—when we have a good RCT to inform a decision that is great, but most times we do not, and we have to take into account many other factors which are hard to weight appropriately.” Those factors, he said, include case series, meta-analyses and even personal experience and anecdotes. Not least, he added, patient’s informed views must be considered.
Saad’s comments here highlight a need for good decision-making even when an RCT is not available, something made clear in a recently -published systematic review published in the Journal of Vascular Access. This examines the “the extent of variation in the planning and recruitment” of RCTs studying the “contentious area” of arteriovenous grafts (AVGs), and its “stark” findings highlight first that there have been only 31 such trials in 31 years, “the vast majority of which exhibited major limitations severe enough to undermine the results”. Its authors, led by David Kingsmore (Queen Elizabeth University Hospital, Glasgow, UK) call for higher-quality RCT and data which echo Saad’s counsel to consider a wider variety of evidence types when coming to a decision.
Making the case for other forms of scientific evidence further, Saad turned to the example of Gerald A Beathard (University of Texas Medical Branch, Galveston, USA), for whom the Beathard award is named. With 579 publications listed on Google Scholar, Saad said he has “contributed an immeasurable amount” to nephrology “without necessarily publishing RCTs”—of that 579, only one publication is an RCT.
Saad recalled asking Beathard how he has been so prolific and was treated to a characteristic “Geraldism” in response. “It is my hobby, I enjoy writing papers,” Beathard said, “I don’t play golf.” Saad punctuated his presentation with several other humorous but pointed quotes. When Beathard trained Saad in surgery, he gave him the following stipulation: “Ted, there are two ways to do this procedure: you can do it my way, or I can do it my way.”
The risks of putting too much emphasis on RCTs for a given clinical decision were also highlighted by Saad, who said such a lack can be “weaponised” in a way that prevents patients getting the best possible care. “We have to make the majority of our clinical decisions without them,” he contended. Saad referred listeners finally to a spoof on RCTs by Robert W Yeh (Harvard Medical School, Boston, USA) published in The BMJ investigating the effects of parachute use on “prevention of death and major trauma when jumping from aircraft”. Not every question, he concluded, can or should be answered by an RCT.
A glimpse of the stars
Elsewhere at ASDIN 2023, Renal Interventions spoke to Beathard himself on the state of nephrology, interventional or otherwise. The prominent researcher made the claim that the creation of vascular access has improved with more widespread training. “The advent of endovascular arteriovenous fistulas (endoAVFs) has me concerned that there is a tendency to over-utilise it,” Beathard said, “but over time I think that will take care of itself.”
A new focus on broader “life planning” for dialysis patients was something Beathard said he welcomed. “When we started, we fixed what was broken when it was broken. Fistula first was an important development in this because it increased people’s awareness of vascular access, leading to the new developments we have seen in training in the field. New ideas are entering the stage, too—the research around drug-eluting balloons (DCBs) is promising, if not quite exciting just yet.”
Asked if he thought the future of nephrology was interventional, he answered yes: “Subspecialties like these will attract new people into nephrology. There are many that did not exist even ten years ago, and it is because of those that young people are attracted to particular fields.” Transplant nephrology and interventional nephrology he cited as examples, with ASDIN training ramping up for the latter. The challenge he outlined, however, is “to train the trainers,” with many programs not having sufficient staff to administer them. His goal at the University of Texas Medical Branch is to meet that challenge, Beathard stated.
“If you can do procedures on your own patients that is a big advantage,” he explained. “General nephrologists today often know little about vascular access, and yet it is so important to dialysis, which is frequently a big part of their practice. They have relegated that to consultants who are sometimes not that well-trained or informed. For the nephrologist to take ownership of that is going to be extremely important.”
“Dialysis patients are very unique. A patient who has a need for a procedure is probably going to have a recurring need for one—having a nephrologist they have an ongoing relationship with, trained to provide that ongoing care, will be vital.” On training itself, he said that he did some things very differently to when he started out in 1985—and even compared to a decade ago, particularly in areas such as excessive blood flow. “Some of those lessons I have learned from the people I am training!”
Finally, he turned to the subject of ASDIN itself, of which he is a founding member: “Meetings like ASDIN are important because I see how others do things there, too, even just in conversation—you learn as much in the hallway as you do in the presentations. You can read papers, but that personal contact is so important. You can ask questions that maybe no one else has. There is no substitute for that.”
“Zoom meetings are okay,” he said finally, coining a new Geraldism, “but it is like the movies: there is a big difference between seeing the stars on screen and seeing them in person.”
