Adoption of Intravascular Imaging Use Among Intermediate to High-Volume Operators in the United States From 2019 to 2023: A Medicare Data Analysis — An Interview With Karthik Murugiah, MBBS, MHS

Dr Karthik Murugiah shares background and insights on his article, “Adoption of Intravascular Imaging Use Among Intermediate to High-Volume Operators in the United States From 2019 to 2023: A Medicare Data Analysis.”

Hello, everyone. I'm Karthik Murugiah. I'm an interventional cardiologist at Yale and a physician-scientist, and I'm excited to join you to talk a little bit about my study.

What prompted your team to study operator-level adoption of intravascular imaging (IVI) rather than simply examining overall national utilization rates? What unanswered clinical questions did you hope this approach would accomplish?

There’s been an increasing amount of data that has accumulated about the benefits of IVI, and that was the trend until this year. Obviously, there's been some mixed data that has come out, but predominantly over the last decade, there was a good amount of data showing that IVI is helpful for percutaneous coronary intervention (PCI) for improvement in outcomes. And it is known that, in the United States, the use of IVI is low compared with other nations. For example, in East Asia, the rates of using intravascular ultrasound (IVUS) or IVI are pretty high, upwards of 80%, and Europe is probably in the similar range, or maybe a tad higher than the US.

The use of IVI in the US has always lacked. We published a study a few years ago in which we showed, at the operator level, that the use of IVI was really low. This is consistent, obviously, at national levels, that we see overall imaging is low, but we wanted to better understand how the distribution of IVUS or IVI use is among operators. Then we noticed that a large proportion of operators do not use IVI or have very low use—over three-fourths of physicians performing PCI did not, or had a really low, use of IVI Also, a very small proportion of the interventionists had a high usage of IVI, about 6% or so.

Four or 5 years have elapsed since then, and we noted in CMS data, or Medicare data, that the use of IVI has been increasing in the US. So, we wanted to understand what kind of operators are adopting IVI, and what kind of operators are not. The motivation for this is, from a policy standpoint, if you want to improve use of IVI, you need to understand whether there are geographical differences, structural differences, or individual operator differences, because that leads to totally different ways of solutions. If there's an individual phenotype of operator that is not using IVI, then we potentially could identify those or target training efforts.

So that was the motivation for this study: we looked at individual operator-level IVI use in 2019 and tracked these operators through to 2023 and tried to see what kind of operators adopted imaging, and what kind of operators did not.

Why do you think some of the busiest cardiologists have been slower to incorporate IVI? Do you think this is more about workflow concerns, just habits, reimbursement issues, or other barriers?

The results are quite thought-provoking, and we sort of always equate higher volumes in PCI with improvement in outcomes, but we see a different result over here with operators who have a high volume of PCI, in fact, using less IVI. So, it is a bit of a surprising result to us. But, from a logistics standpoint, I can imagine, and I can totally understand, if someone is doing a really high number of PCI, it does add some degree of time, because it's not just acquisition of the imaging, you have to interpret the imaging, make some decisions based on that. And these operators, maybe just to have throughput and get done with the day, maybe choose not to use imaging.

The other aspect that you mentioned was about compensation. I'm not sure how much that affects it. Yes, there may be certain differences in contracting, etc, for individual hospitals, so that may play a role, but we shouldn't see a national signal in that. So, it is quite interesting why high-volume operators use less imaging, and it was a surprising finding in that sense.

You also found that once operators adopted IVI, very few reverted to low use. What do you think this says about the clinical value of IVI once they've incorporated it into their practice?

It’s not necessarily that these people adopted IVI during the period—it's more like tracking the people who were already using high IVI in 2019, and that they continued to do so in 2023. I think it's just a matter of what kind of imaging modality you're comfortable with, and if someone really relies on IVI for sizing and all kinds of parameters before deploying stents, then I think that person is not really going to be changing their practice.

In the same vein, we also noticed that operators who newly entered practice seemed to use high rates of IVI; again, it would be interesting to see how these operators continue to use imaging down the road, but it could also be that that is how they were trained to do PCI. So those are a couple of factors that potentially could be responsible.

There is a growing body of evidence supporting image-guided PCI, and with all the recent guideline updates, what do you think are the most important clinical implications of your findings, and how should hospitals, training programs, and cardiologists use these data to improve patient care and reduce the disparities in IVI adoption?

In terms of evidence, this study was tracking operators from 2019 to 2023. The large trials during this time—for example, the IVUS-XPL trial in 2015, the ULTIMATE trial in 2018, and also the RENOVATE-COMPLEX-PCI trial in 2023—all of them showed benefits of IVI, and that led to a series of guideline changes in 2021, and now in 2025, with IVI getting a 2A recommendation and a 1A in the ACS guidelines. So all the evidence points out that IVI is helpful.

Recently, though, there has been some mixed evidence with the OPTIMAL trial and IVUS-CHIP trial, so it'll be interesting to see how that affects IVI. But, putting that aside about the benefits of IVI, I think that we still need more information as to what kind of cases it's helpful in and what kind of cases it is not.

But to improve IVI, the details that have emerged about certain regional differences are helpful; certain types of operators who've been in practice for a long time may benefit from some sort of outreach or training possibilities, perhaps through our societies, that could focus on these kind of operators and have structured programs which could give them all the knowledge they need to use it effectively.

So those are some of the implications of the program. And IVI also could be measured as a performance measure, etc. I think these will be some changes that would drive adoption in the future.

Is there anything else you'd like to add for our audience?

Obviously, there's some caveats to our study because the CMS data does suppress less than 10 PCI, so that is why we had to focus on intermediate and high-volume operators. But I think the signals that emerge from the study are still quite reliable, despite these restrictions to what kind of operators we studied. These are robust signals and are thought-provoking, so I hope they spur the science to see how technologies are diffused and adopted in cardiology. So, I think in that way it’s a thought-provoking article. I thank you for highlighting this study and chatting with me about it.

The transcript has been lightly edited for clarity.