The results of the BEST-CLI trial, which were published in the New England Journal of Medicine, yielded surprising results regarding the role of endovascular therapy in the treatment of critical limb ischemia. The data presents a significant difference between endovascular therapy and traditional bypass surgery with respect to major adverse limb events or reintervention.
On Dec. 6, SIR hosted a members-only town hall to discuss the trial, review the data and consider how the results will impact IR practices.
What is BEST-CLI?
Best Endovascular vs. Best Surgical Therapy in Patients with Critical Limb Ischemia, or BEST-CLI, is an international prospective, randomized research study aimed at determining the best revascularization treatment for patients with critical limb ischemia. The study was funded by the National Institutes of Health and is a follow-up to the BASIL study which was published over 15 years ago in The Lancet. BASIL randomized patients for bypass or endovascular surgery, and after 2 years follow-up found no difference in the endpoints of overall survival or amputation free survival.
Unlike BASIL, BEST-CLI researchers judged their primary outcome by the incidence of major adverse limb event (MALE) or death of any cause, and their secondary outcome was major limb events or post-operative death within 30 days, as well as minor interventions.
Over 5 years, 1830 patients at 150 sites were enrolled and divided into two cohorts, patients with a single segment of greater saphenous vein (SSGSV) as a potential conduit were enrolled into cohort 1. Those without a SSGSV were enrolled into cohort 2.
Results
According to Robert A. Lookstein, MD, FSIR, who presented during the town hall, the BEST-CLI authors reported that among patients who had adequate SSGSV for surgical revascularization (cohort 1), the incidence of a major adverse limb event or death was significantly lower in the surgical group than in the endovascular group. Among patients who lacked an adequate SSGSV conduit (cohort 2), the outcomes in the two groups were similar.
Within cohort 1, the rate of major adverse limb events or death was 42.6% in patients who received bypass, compared to 57.4% in patients who underwent endovascular therapy. Of those who received surgery, 9.2% had reintervention compared to 23.5% of endovascular patients. Of note, the acute technical success rates of the revascularization procedures were discordant—98.3% of the surgical procedures were successful, whereas only 84.7% of the endovascular reinterventions were successful.
“If you look at the curves for outcomes, the only real difference between bypass and endovascular surgery is in the early reintervention rates,” said Sanjay Misra, MD, FSIR, who was a site investigator for BEST-CLI. “If this study were set up like BASIL and looked primarily at major amputation, the BEST-CLI results would show equivalent outcomes.”
According to Drs. Misra and Lookstein, the technical failure rates are evident within the first 90 days of data, which show high early major reintervention rates for endovascular therapy.
“Every one of us knows when we’re going to fail or have a reintervention. We can tell on our table, and for most of us, we will then reattempt, or follow the patient,” Dr. Misra said. “I believe that a better-balanced assessment would be to look at successful endovascular outcomes and compare them to the successful surgical outcomes.”
Technical success rate
The technical success rate of 84.7% is surprising, but the parameters defining failure are key, Dr. Lookstein says. For the surgery group, occlusion of the bypass graft or failure to achieve a patent bypass graft at the completion of the procedure constituted a technical failure. For endovascular surgery, a failure was determined by inability to cross a stenosis or occlusion, or a residual obstruction of more than 50% in the superficial femoral artery, popliteal or all tibial arteries, such that there was no in-line flow to the foot.
Panelists agreed that the low success rate doesn’t seem accurate. Dr. Misra pulled three papers from the previous decade’s endovascular literature that showed technical success rates as high as 89–93%.
“I feel that anything below a 95% success rate is unacceptable,” said panelist Kumar Madassery, MD. “Whatever vascular practice you’re in, that success rate isn’t good.”
Dr. Lookstein also pointed out the breakdown of who performed these procedures.
“Almost the entire investigator list for cohort 1 were vascular surgeons,” he said. “IRs made up a very small number of the investigators.”
Panelist Bret N. Wiechmann, MD, FSIR, agreed. “There is an high crossover rate within the first 90 days. How many IRs wouldn’t give it another shot if we fail the first time?”
The endovascular data shows contemporary, advanced technologies and techniques largely weren’t used—most patients were treated with plain angioplasty alone. Drug coated balloons and stents were rarely used and there were low rates of atherectomy or alternate access—Dr. Misra points out that the technical failure rate is closer to data from almost 20 years ago.
“A success rate of 84.7% is not what we would expect, and it raises a lot of questions,” he said. “What was the TASC classification on the failures? Were there reattempts on the table?”
The lack of TransAtlantic InterSociety Consensus (TASC) classification data makes it difficult to understand the factors that led to the high technical failure rate, Dr. Misra said. Referring back to the earlier papers he pulled, all listed procedural details, including TASC classification.
Missingness
The TASC classification is among several other points of data that haven’t been released from BEST-CLI, which have contributed to a higher rate of missingness than peer NIH studies such as ATTRACT (10.5%), CORAL (11.2%) or CLEVER (8.3%).
“Doing research in the United States today is very hard,” Dr. Misra said, noting the hard work and skill of the lead researchers and all investigators. “But this is a high rate of missingness—one you could accept for retrospective, but not at this level.”
Dr. Misra urged IRs to compare the BEST-CLI rates to their own data—if IRs have access to it.
“How many of us know our own data and outcomes? How many of us are keeping logs?” he asked. “We have to support our own data sets, to show that an 85% success rate is not good enough and not accurate. We have to support our own registries, like VIRTEX.”
Patient demographics
Drs. Lookstein and Misra agree that IRs should look closely at the BEST-CLI patient demographics and exclusion requirements to see how generalizable the results are.
“When you look at the cohort breakdowns, you see that only 28% are women, 20% are Black, 13% are Hispanic and only 10% have renal insufficiency or renal failure,” Dr. Lookstein said. “You have to wonder, how applicable is this to the general population, and to my patients specifically?”
Dr. Lookstein focused on the analysis and confidence interval data within the trial, which indicated no benefit of bypass over endovascular surgery for Black patients, patients over the age of 80, or those with existing renal dysfunction—which, as many of the town hall panelists pointed out, are key demographics which many IRs serve for CLI treatment.
“Remember that patients in the trial had to have a good vein and be a good surgical candidate,” Dr. Misra said. “Those are fairly restrictive inclusion criteria.”
According to Srini Tummala, MD, FSIR, most of the patients he sees in South Florida would not qualify for BEST-CLI due to having had a coronary artery bypass graft, or just not having a suitable vein.
Parag Patel, MD, FSIR, who also served as a site investigator for BEST-CLI, agreed. “The cohorts don’t fully represent the patients we usually see,” he said. “Our patients have high rates of diabetes and chronic kidney disease.”
Overall impact
Though the study has raised multiple questions, the largest is “What impact will the data have on IR practices, referral patterns and relationships with vascular surgeons?” According to Mary Constantino, MD, FSIR, a panelist at the town hall, it likely won’t impact her own referrals or patient selection.
“I work closely with the vascular surgeons in town,” she said. “In my practice, I’ve started to do vein mapping on everyone and for patients with good veins, high platelets and small vessels, I send them directly to surgeons first.”
Dr. Lookstein also doesn’t expect to see changes at his institution.
“We have had meetings and spoken openly with our vascular surgeon colleagues, whose endovascular success rate is higher than this trial,” he said. “We agree that you need to look for the best revascularization specialists—not just vascular surgeons. If you’re an endovascular expert with a 95% technical success rate, you will have excellent outcomes for your patients. So IR and our vascular surgery group have reached the consensus that this trial should not apply broadly to the entire US population suffering from CLI.”
What comes next
There’s still data to unpack, such as the BEST-CLI registry, which has not released results yet. Without that and other data, Dr. Misra says, it’s hard to plan another study or think about what to question next.
However, all panelists agreed: more data, delivered by IRs, is needed.
“We need to make sure the best data gets out there, and if we don’t do it ourselves, we’ll have to rely on trials like this,” Dr. Lookstein said. “If IRs want better data, we need to get involved in prospective research and produce it ourselves.”
