Two Neutral IVUS Trials in Complex PCI—and One Positive—Spark Debate

NEW ORLEANS, LA—Three trials comparing IVUS with angiography for PCI guidance in patients with complex coronary artery disease took many attendees of American College of Cardiology (ACC) 2026 Scientific Session by surprise because only one, DKCRUSH VIII, showed there was any advantage to using intravascular imaging.

In patients with true complex bifurcation lesions, IVUS guidance reduced the rate of target-vessel failure through 1 year compared with angiography (HR 0.40; 95% CI 0.23-0.71), driven by lower rates of target-vessel MI and clinically driven TVR, Shao-Liang Chen, MD (Nanjing First Hospital, China), reported during the late-breaking clinical trials session.

DKCRUSH VIII was sandwiched between two European trials that delivered neutral results. In OPTIMAL, a study of patients with unprotected left main disease, and IVUS-CHIP, which included a mix of patients undergoing complex high-risk PCI, IVUS guidance did not improve overall clinical outcomes compared with angiography alone.

Experts wrestled with the findings, which conflict with accumulated evidence demonstrating the benefit of IVUS during difficult PCI cases and strong guideline recommendations that support intracoronary imaging. For instance, in the latest US ACS guidelines released last year, there is a class 1A recommendation to use IVUS or OCT to guide PCI in patients receiving a stent in the left main artery or in complex lesions.

“There’s a lot of controversy that’s introduced by these presentations, and it’ll be really interesting to see how this informs our practice,” interventional cardiologist Douglas Drachman, MD (Massachusetts General Hospital, Boston), told TCTMD, adding that he uses intravascular imaging in at least 95% of his cases.

Much of the discussion of the results centered around the extensive IVUS experience possessed by operators who participated in the trials and whether that skill informed and enhanced the techniques used during cases when angiography alone was used. Some physicians questioned whether these results would be applicable to broader practice across less-experienced centers.

“They may be able to glean from what they would ordinarily see using an intravascular ultrasound catheter what would need to be done, and they just do it automatically,” Drachman, a member of the ACC Interventional Section, said. “It’s hard when your control group performs above what happens in the real world to demonstrate benefit from the test group.”

It’s hard to disprove rigorously conducted science in a randomized controlled trial and say it can’t be. Douglas Drachman

Still, these are high-quality trials that provide new insights into the use of IVUS in specific subsets of patients or lesion types, Drachman said.

“We have to really explore this,” he said. “I don’t want to be trapped by dogma and just say that I so firmly believe that it has to be the case [that IVUS is beneficial] when there are now new data that influence our perspective. I think we’re going to have to look really carefully at these, because it’s hard to disprove rigorously conducted science in a randomized controlled trial and say it can’t be. But I think we have to understand how. How can it be? And that will help us to unravel the story a little bit better and then inform the guidelines.”

In the meantime, Drachman urged, “It’s really important that we continue to advocate that people become comfortable using intravascular imaging and [position it as] a default strategy, because I think it elevates the quality of the care that we deliver.”

In Japan and other parts of Asia, 85% to 90% of PCIs involve IVUS guidance, much higher than the 15% to 20% of procedures guided by IVUS in the United States, he noted.

DKCRUSH VIII

DKCRUSH VIII, published simultaneously in JACC, was conducted at 24 centers in China. It included patients with true complex bifurcation lesions with a side-branch lesion length of at least 10 mm and diameter stenosis of 70% to 90% plus two other minor criteria: moderate-to-severe calcification, multiple lesions, bifurcation angle < 45° or > 70°, main vessel reference vessel diameter ≤ 2.5 mm, thrombus-containing lesions, or main vessel lesion length ≥ 25 mm.

Researchers randomized 556 patients (mean age roughly 67 years; 23% women) who were undergoing PCI with the double-kissing (DK) crush stenting technique to IVUS or angiography guidance. In the IVUS group, imaging was used throughout the procedure, including in 98% of patients before PCI and in 92% after final proximal optimization. At the end of the procedure, optimal IVUS criteria were met in 75.4% of patients.

The primary endpoint was target-vessel failure, a composite of cardiac death, target-vessel MI, or clinically driven target-vessel revascularization. Through 1 year, the relative rate was 60% lower in the IVUS versus angiography group (6.1% vs 14.7%). There were significantly lower rates of target-vessel MI (4.3% vs 9.4%; HR 0.46; 95% CI 0.23-0.90) and clinically driven TVR (2.9% vs 7.6%; HR 0.37; 95% CI 0.16-0.84).

The rate of the primary endpoint was particularly low for patients who met optimal IVUS criteria (2.6%), whereas the rate was similar to what was seen in the angiography-guided arm (15.9%) when optimal criteria were not met.

OPTIMAL

Luca Testa, MD (IRCCS Policlinico San Donato, Milan, Italy), presented the results of OPTIMAL, which was published simultaneously in the New England Journal of Medicine.

Conducted at 28 centers in Italy, Spain, and the United Kingdom, the trial included 806 patients (mean age 71 years; 22% women) with unprotected left main disease. The mean anatomical SYNTAX score was about 30.

Patients randomized to IVUS versus angiography guidance had more complex coronary disease as indicated by a higher prevalence of class B2/C lesions and greater use of lesion preparation devices, including compliant and noncompliant balloons, cutting balloons, Rotablator, and intravascular lithotripsy.

The use of the proximal optimization technique was similarly high in both the IVUS and angiography groups (89.6% and 85.1%, respectively). Total procedure time was longer in the IVUS arm (88.6 vs 63.9 minutes), which is consistent with use of IVUS leading to additional maneuvers, Testa said.

The primary endpoint was a patient-oriented composite of all-cause death, any stroke, any MI, or any revascularization. At a median follow-up of 2.9 years, the rate was 33.7% in the IVUS arm and 30.9% in the angiography arm (HR 1.11; 95% CI 0.87-1.42). Results were similar for a device-oriented endpoint encompassing CV death, target-vessel MI, and clinically indicated TLR and for a vessel-oriented endpoint consisting of CV death, target-vessel MI, and TVR.

Testa noted that at least one optimization criterion was not met when IVUS was used at the end of the procedure in 27% of cases. But, he said that even in that subgroup, “the numbers were absolutely acceptable in terms of what we established as the reference.”

I just say: be careful with the interpretation of the data. Juan Granada

Overall, the OPTIMAL results indicate that within high-volume centers and in the hands of operators with IVUS expertise, there is “a potential recalibration of the angio-based measurements and evaluation following that large experience with IVUS,” Testa said.

Juan Granada, MD (Cardiovascular Research Foundation, New York, NY), in his discussion following Testa’s presentation, agreed that “we need to be careful about the interpretation of the data” in light of the skill of the operators involved. “For me, the main message of the study is that a high-volume, skilled operator can achieve comparable results when using IVUS or . . . using angiogram alone, not necessarily that [it] can be extrapolated [to] the general population.”

The use of IVUS, especially in complex cases, continues to be “highly warranted and recommended,” Granada said.

“I always suggest to use IVUS in all situations where something’s not clear,” said Testa. “Obviously, it is also really related to years and years of usage, years of experience, to reach a certain level of confidence where you can have the choice of deciding whether it’s mandatory or it’s just a plus. For everyone else, I just say be careful with the interpretation of the data.”

IVUS-CHIP

IVUS-CHIP, presented by Roberto Diletti, MD, PhD (Thoraxcenter, Erasmus University Medical Center, Rotterdam, the Netherlands), and also published in NEJM, was conducted at 37 centers in seven European countries. It included 2,020 patients (mean age 69 years; 20.6% women) who were undergoing PCI for complex coronary lesions, which encompassed those with angiographic heavy calcification (more than 40% of patients), ostial lesions (about 27%), true bifurcation lesions involving side branches > 2.5 mm (about one-third), left main artery lesions (about one-fifth), chronic total occlusions (22%), in-stent restenosis (17%), and long lesions with an estimated stent length > 28 mm (about 60%).

As in OPTIMAL, mean procedure duration was longer in the IVUS group (88.8 vs 66.2 minutes).

The primary endpoint was TVF, a composite of cardiac death, target-vessel MI, or clinically indicated TVR. Through a median follow-up of 19 months, events were numerically, but not significantly, higher in the IVUS versus angiography arm (13.9% vs 11.1%; HR 1.25; 95% CI 0.97-1.60), with similar trends seen for each individual component of the endpoint.

IVUS held the advantage for definite/probable stent thrombosis (0.5% vs 1.5%; HR 0.33; 95% CI 0.12-0.90) and definite stent thrombosis (0.2% vs 1.0%; HR 0.20; 95% CI 0.04-0.90).

We are reducing the rate of underexpansion, which is the most important independent predictor of stent thrombosis. Roberto Diletti

Discussing the results after Diletti’s presentation, Deepak Bhatt, MD (Icahn School of Medicine at Mount Sinai, New York, NY), said the stent thrombosis finding is an important one, despite being a secondary outcome, adding that it would be sufficient to justify the use of intravascular imaging during PCI in this population of complex patients.

Diletti said there is a mechanistic rationale to explain a lower risk of stent thrombosis after using IVUS, pointing to a likely improvement in stent expansion. “We are reducing the rate of underexpansion, which is the most important independent predictor of stent thrombosis, so this could be an important point to consider imaging when doing PCI, especially complex PCI,” he said.

Sorting Through the Discrepant Results

Several physicians agreed that it’s possible that the high level of IVUS expertise by operators in the trials could explain the neutral results of OPTIMAL and IVUS-CHIP, whereas the adjusted stenting techniques with IVUS in bifurcation lesions could explain the positive result.

DKCRUSH VIII “continues to build on our growing literature and evidence base, suggesting that the use of intravascular imaging and meticulous attention to detail in complex coronary substrates can be beneficial in improving outcomes,” Drachman said at a media briefing.

Ajay Kirtane, MD (NewYork-Presbyterian/Columbia University Irving Medical Center, New York, NY), discussing the results after Chen’s presentation, said a unique aspect of the trial was the technical differences discovered using IVUS and how that modified the PCI procedure.

“This is a pretty technically involved procedure, the DK crush stent, and you were able to look through the specific areas of the technique that were changed based upon the imaging,” he said, “And so, to me, it’s somewhat unsurprising then that you were able to demonstrate a difference, especially with where the wire was located and how that impacts the geometry of the stents.”

To explain the neutral results of the other two trials, Drachman said certain aspects of the methodologies, patient populations, and clinicians likely contributed. In OPTIMAL, for instance, there appeared to be increased lesion complexity in the IVUS versus angiography arm, a possibly bulky device was used, and there were infrequent changes in approach in response to the IVUS findings.

As an operator, Drachman said IVUS reveals something that can be touched up nearly every time it’s used and those modifications ultimately provide benefit. He didn’t get the sense that IVUS altered techniques as often in OPTIMAL and IVUS-CHIP as in DKCRUSH VIII.

“If you’re just taking pictures of something and then not modifying what you do, it’s probably not going to have any impact,” Drachman said.

Continued Advocacy for IVUS

When using IVUS, Drachman said information is gained that can’t be judged with angiography alone and this is particularly important as the complexity of patients continues to increase.

Intravascular imaging “helps you identify if there’s calcium. It helps you identify when there’s diffuse plaque, or you may not see it as diffuse as it is when you perform an angiogram. It helps identify the size and caliber of the vessel as well as the composition of the plaque. These are really critical things for making decisions of how to treat patients,” Drachman said.

IVUS is a good teacher and it gives you a lot of information. It gives you a lot of training. Luca Testa

Testa said, “I believe that IVUS is a good teacher and it gives you a lot of information. It gives you a lot of training.”

Diletti, too, touted the importance of learning how to use IVUS. He suggested that early on, operators should learn how to do PCI with angiography guidance and IVUS support. “IVUS can inform us much more than angio on what we are doing,” he said. “And of course, when we become very skilled in doing PCI, probably we can use IVUS in a selective part of our patients, not in all patients.”

“I would hope that findings of these types of studies don’t allow people who never have performed intravascular imaging to say, ‘I don’t need to learn it. I don’t need to use it for my patient,’” added Drachman. The studies suggested “you do need to use it, but there may be a career stage in which you’ve used it so often that you can do all of those steps automatically as you do with the guidance of intravascular imaging.”