Guidelines for Peripheral Interventions Often Built on Shaky Foundation

Evidence underlying the American Heart Association/American College of Cardiology (AHA/ACC) guidelines for peripheral vascular intervention (PVI) is often of poor quality, a new analysis has found.

The fact that much of the data are derived from observational studies speaks to the need for more randomized controlled trials in the field, emphasized lead author Partha Sardar, MD (Brown University, Providence, RI).

“We need to continue to generate the evidence that’s needed,” and seek additional research funding from government and industry, said Sardar. The paper, published this month in Circulation: Cardiovascular Interventions, is authored by many of the leading names in the field who, according to Sardar, undertook the analysis after noting the relative dearth of PVI studies informing this space.

Co-author Michael R. Jaff, DO (Newton-Wellesley Hospital, MA), observed that the findings are more dismal than he would have predicted at the outset. “But it doesn’t surprise me, because we’ve been struggling with quality data on PAD interventions for as long as I’ve been in this field,” he told TCTMD.

The lessons contained in the current analysis represent a “cry for holding people to a higher standard,” Jaff said, adding that “it’s time to step up the game” in terms of study design. There is good reason to believe expectations are rising and, what with randomized trials like IMPERIAL and others being conducted in the peripheral space, already being applied, he said.

As pointed out in the paper, several registries were launched in recent years, including the Society for Vascular Surgery’s Vascular Quality Initiative and the National Cardiovascular Data Registry-PVI Registry.

This is good news, but observational studies are no substitute for an RCT, Jaff urged. Outcomes aren’t independently adjudicated, and there’s the potential for less accuracy in reporting. “Registries are good, don’t get me wrong. They give us knowledge of what’s happening in the field in real-world situations. There’s great value to that. But it’s not level of evidence A,” he said.

Sardar concurred. Registries are “a good way to start,” he said. “At the same, we need more randomized data to support practice.”

Just last month, a meta-analysis of 28 RCTs uncovered an unexpected signal of late death with paclitaxel balloons and stents in PAD, prompting BASIL-3 and SWEDEPAD investigators to halt recruitment for their trials until more is known.

It’s time to step up the game. Michael R. Jaff

Sardar and colleagues turned to five current AHA/ACC guideline documents that covered the areas of endovascular and surgical PVI, the most recent being the 2016 lower-extremity PAD recommendations. They analyzed both the level of evidence and the strength of recommendations.

Altogether, the guidelines contained 134 recommendations: 13% were backed by level A evidence and 39% by level C evidence. Approximately half (54%) were class II and one-third (35%) were class I.

The situation was particularly dire for interventions related to pulmonary embolism/deep vein thrombosis, inferior vena cava filters, and renal artery stenosis, none of which had any level A evidence supporting the recommendations.

In contrast, 24% of endovascular stroke therapy recommendations were supported by level of evidence A.

Over time, the total number of recommendations for treating lower-extremity PAD decreased, though there was no shift in the quality of evidence.

What’s Holding Back the Peripheral Field?

Asked why the evidence base for PVI is so slim in comparison to cardiac procedures, Jaff replied, “First of all, if you look at coronary stents . . . at each step along the way there was really only one specialty involved: interventional cardiologists. And they had one, maybe two societies: ACC and [the Society for Cardiovascular Angiography and Interventions]. So it was pretty easy to come to agreement on trial design, populations to study, endpoints and definitions of endpoints, duration of follow-up, and levels of rigor.”

But when various specialties are all working in one area, as is the case here, and they’re in competition, he continued, “it becomes a race: who’s going to put their stake in the ground faster—cardiologists, surgeons, radiologists, medical vascular specialists? It really has been decades of competition. And as a result of that, each specialty retracts into itself.”

Multidisciplinary guidelines and, most recently, the appropriate use criteria for PAD are “obvious exceptions,” Jaff added, as is the Peripheral Academic Research Consortium’s collaborative effort to standardize definitions and outcomes. “These are all steps in the right direction,” he commented.

Another factor is that only recently have there been head-to-head comparisons in randomized trials of different peripheral devices. “When all you have are registries—‘I’ve put in this number of devices in this number of patients and these are the results I got’—it’s not high levels of science,” he commented.

Sardar, for his part, said that patient recruitment might be difficult in the peripheral arena, as there are fewer cases than in the cardiac space. Moreover, there is comparatively less interest from manufacturers and the US National Institutes of Health in funding PVI studies, he added.

How to Cope, and Move Forward

What should you do, when even the guidelines aren’t definitive? Try your best, Jaff advised.

“You practice based on what the literature says to the best that you can. That’s the great part about being a doctor: you still use your judgement because you’re trying to do what’s best,” he explained. “So when the literature doesn’t give you an answer, when the evidence level is C, then you look at the patient and say, ‘Look, the literature really doesn’t guide us, but in your specific situation, here’s what I think we should do based on what I know about you and my experience and expertise.’ And I have to hope that that’s what the physicians are doing.”

When reading guidelines, clinicians should look carefully for details on the level of evidence and be aware of variations in quality across PVI subspecialties, Sardar suggested. “When the level of evidence is high, they can strongly, confidently tell their patients [whether] there’s going to be benefit or harm. But in certain areas definitely there is no clear benefit or harm based on observational data, and maybe what we are doing isn’t supported by strong evidence.”

David W. Lee, MD, and Matthew A. Cavender, MD (both of the University of North Carolina at Chapel Hill), point out in an accompanying editorial that peripheral vascular disease, particularly PAD, “remains underdiagnosed and undertreated despite efforts to identify those patients at high risk and an expanding number of lifestyle, pharmacological, and interventional therapies available for use in [treatment].”

As such, the standardization of treatment pathways is especially needed, they say, calling for comparative-effectiveness research to guide this effort.

“Improvements in the quality of comparative effectiveness form the foundation upon which clinical practice guidelines, the development of performance measures to help identify areas of deficiency and success, and initiatives to improve the quality of healthcare delivery are built,” Lee and Cavender write.