us Interview with Giulio Guagliumi, MD
Most Popular Angioplasty Web Site
supported by Volcano Corporation

Giulio Guagliumi, MD

Dr. Giulio Guagliumi, an internationally recognized leader in the use of Optical Coherence Tomography (OCT) in the coronary arteries, practices in the Cardiovascular Department of Ospedali Riuniti di Bergamo in Bergamo, Italy. He has spoken at numerous conferences worldwide and recently brought together experts for a course in Bergamo, titled "OCT, IVUS, FFR: When, Where, How" -- in which the clinical applications of these technologies were explored and demonstrated.

Dr. Guagliumi has been published in numerous peer-reviewed journals and is principal investigator in several clinical trials assessing the performance of stents inside the coronary artery, using OCT imaging. One recent study, the ODESSA Trial (Optical coherence tomography for DES SAfety) was presented at TCT 2008 and was one of the first looks at drug-eluting stent coverage and healing inside the artery. OCT imaging, although a very new modality for interventional cardiologists, is now being utilized by the medical profession and industry during the early stages of stent design to assess safety issues of strut coverage.



Giulio Guagliumi, MD
Giulio Guagliumi, MD

Q: Much has been written about intravascular ultrasound IVUS and its usefulness. But optical coherence tomography (OCT) is the new kid on the block. Where is this technology right now, with regard to both research and clinical application?
Dr. Guagliumi: When you are working in a new field, opening a new window, you have a lot of expectation and you need time to realize the level you can reach in clinical intent. These are spectral technologies. IVUS has been with us for 20 years -- OCT is more recent -- but in either case we are talking about something that is quickly changing.

OCT image from ODESSA Trial
OCT image from ODESSA
Trial, showing covered and
uncovered struts of a
drug-eluting stent

We started to use OCT in the coronary field a couple of years ago. The first group started to use it to detect vulnerable plaque. Then, and I think this is one of the merits of my group, we quickly switched to something that was dramatically more important in clinical use: "Can we derive information on the coverage of the stent struts and the healing process?" So we started working with the problem of late and very late stent thrombosis.

But during the last couple of years this was done with early OCT technology that was quite difficult to use, because you had to be quite expert: you had to occlude the vessel and flush away the blood, just to allow the fibers to have a perfect clean vision of the vessel. Because, of course, it's a light technology, so the blood is going to create the problem. So that was done systematically with very dedicated work for a couple of years. And you needed a very skilled team, just to have the best images without artifacts and contamination.

Today, two years later, we are discussing the possibility of scanning an entire coronary artery in 3 seconds and obtaining high quality images without needing any occlusion. That’s what we showed at our course. So it's not just one technology. It's something that is changing extremely quickly. And the level of the clinical information that we can get is extremely dynamic. We have a lot of expectation. It is the beginning of the story.

Just comparing OCT with IVUS today, with the current technology we have at least a 10-30 times higher resolution. We can analyze one stent not in a few sections, but in hundreds of sections. The problem is: can we deal with all these data very quickly and very appropriately without anything in excess? And that is what's going on -- but a very dynamic situation with a lot of potential improvement just around the corner. We need time.

Q: Can some of these technologies be used clinically right now? Certainly IVUS is being used much more than it was. OCT, as you're saying, is more of a research tool right now.
Dr. Guagliumi: This new technology, where you are able to scan one coronary artery in 3 seconds, means that you can do this without a super-expert team. And that could be extremely important. Even today we are just normally performing many of these evaluations in stented patients to determine healing. There are papers coming, probably in one or two months, showing that the correlation with histology for coverage is very very high -- quite impressive. So it's going to change the level of understanding in a practical way of the stent-vascular response.

I can say that when we are discussing late and very late stent thrombosis, people have in mind that it's just one single phenomenon. But it is something that is very complex. Sometimes when you are using this technology to cut the stents in slices of cross-sections, in vivo, you may see four different types of responses in only 4mm. So you might have a beautiful stent for 95% and then you can run into trouble in 2 or 3mm. And that is going to be extremely helpful to guide us. Number one, to throw out the most toxic technology and two, all types of investigation in the next generation of stents is going to include this technology as an early evaluation before going into thousands of patients with the final release – so in just a small group of patients there is good information about the healing process.

What about the clinical practice every day? We have already shown that OCT is a very sensitive technology for thrombus detection, for malapposition -- much more accurate. We still have to prove that all of this information is going to impact the outcome.

Q: With the Horizons-AMI OCT subset there were some comments, for example, by Dr. Renu Virmani that you couldn’t really tell what types of cells were covering the struts: they might be fibrin and not endothelium.
Dr. Guagliumi: Of course this is true and it is something we need to target. As I said before, this is the first step. Light is the most flexible radiation available. You can use not only the intensity, you also can use polarization, and you can use spectroscopy to determine the chemical aspect -- that is the next target. We have already started with tissue characterization. Until now we used just one single parameter: the intensity of the signal. And now we are opening a new phase. But what is important is the first step: can we scan the entire coronary artery in a short period of time without asking the people who are doing it to be extremely expert? Now we can say yes, because now it's a simple procedure. The next step is the separation between fibrin and neointima -- and then we are going to something even more precise, probably with spectroscopy.

Q: You've said that the data coming from these studies may impact the design of new stents and other new technologies?
Dr. Guagliumi: This concept has been accepted by the European Society of Cardiology. A little over a year ago, we had kind of a consensus meeting, which has been published in the European Heart Journal, EuroIntervention. On the team was Eugene Braunwald, Patrick Serruys, Mitchell Krucoff and many others, including a web transmission from the U.S. We discussed what is important for the DES application. And one of the written conclusions for the future was that we wanted to have more than only QCA data -- although QCA data are very reliable for restenosis, they are not telling you anything about the healing process. So the paper recommends that for the future generations of DES, we should have systematically collected IVUS data and OCT data and endothelial functional data. That is a sort of an official statement, and the FDA is going to follow the same pattern.
    Giulio Guagliumi in cath lab
Dr. Guagliumi in cath lab

Q: This would be for approvals in the future?
Dr. Guagliumi: And for checking the level of risk before putting devices into thousands and thousands of patients.

Q: One interesting outcome from a trial that you did, the ODESSA Trial, looked at the four different drug-eluting stent types and came up with some pretty interesting data about the differences among them.
Dr. Guagliumi: First of all, it is confirmed that DES are not just one single family. Each one has its own profile. Logically the more you push the envelope in trying to make zero neointimal growth, the more you risk leaving some struts uncovered, or some struts malapposed. So OCT technology is going to dramatically increase the understanding of the response. As I said before, don’t think about the stents having an all-or-nothing response. It is a very sophisticated response from any different site of a single stent. So it's a spectrum and it depends upon the interaction between plaque and drugs, concentration, but for sure these stents are going to be extremely different and you can see these differences by using this technology.

Q: One striking finding was that strut coverage was much greater on the Endeavor zotarolimus-eluting stent than with the others.
Dr. Guagliumi: The data fit with all the scientific data previously published. When you have a late lumen loss on the average of 0.6 for the zotarolimus stent, you have more tissue. But the stronger message is not that you have more tissue -- it's that you have a very uniform response. So zotarolimus is very homogeneous, and I can guarantee that, having seen some very recent graphs.

We also just finished another study -- named OCTAMI. It is the same use of zotarolimus vs. bare metal Driver in acute myocardial infarction -- so a different setting, just to see if the same type of uniform behavior and coverage is maintained in maximal critical situations such as AMI.

For the other stents, as I said, the concept is heterogeneity. So the more you are aggressive, the less you have this uniform response. You might have great section with nice coverage, but you might have a section that is completely devoid from coverage. And that's in fitting with the value of the late lumen loss around zero. So the problem is how we can get the nice balance, not too much and not too thin.

Q: Have you been able to definitely link uncovered struts with increased stent thrombosis?
Dr. Guagliumi: Two points. Number one, all the patients in all these studies are clinically followed up. Just to see the relationship between the images and the clinical outcome. But in this case we use the pool of all patients, so we are following ODESSA, we are following Horizon-OCT, we are following OCTAMI, we are following OCT-DESI, we are following LONG-OCT, we are following many, many trials. We need enough numbers.

Another critical point is we need a high quality of images and that is my major preoccupation. If you have this type of heterogenetic response, it's not easy to get great images for all the sections, all the stents, all the patients. And if you are going to deal with the artifacts, with suboptimal images, perhaps you cannot be so precise. So we are just trying to put on the paper the concept of the artifacts, how to try to be as clean as possible and to have the most reliable data.

But, with hundreds of these patients followed clinically, and we are already at two years, at the end we will arrive to see if there is a relationship. I can anticipate something. I wrote an editorial for JACC: Interventions for a paper that was published in May by a Japanese group, working with Stanford. For the first time they showed that uncovered struts are related to thrombus formation.

Angioplasty.Org Image
IVUS Study, courtesy Volcano Corp.

Q: What about the role of other intravascular technologies, such as IVUS?
Dr. Guagliumi: Just to give you an idea, my last talk at the ACC during the official OCT meeting was titled, "IVUS is dead." The title was not my choice! But it gives you the feedback that when you are pushed, you start to react.

So what we can expect is a significant improvement in terms of IVUS technology. We already have a new generation of catheters that can be tracked easily in the distal part of the coronary arteries with more pushability, more lubricious. That is one. Then we are going to have a much quicker pull-back. We are going to have a higher numbers of frames.

My idea is that, in any case, light is extremely superior to ultrasound, to sound waves but, of course, light has limitations such as penetration. Light has difficulty in going deeply into the tissue. That is the reason why in all of the studies and all the patients, we used both technologies in pair. OCT just to get the information from the internal surface of the vessel which is much more detailed, but also IVUS to get information about the outside part of the vessel, remodeling for instance. So this type of combination is the true projection for the future, because we think that for a certain period of time, light will remain limited in deeper tissue penetration.

Q: So a system that allows you to use both OCT and IVUS is helpful.
Dr. Guagliumi: Yes. There are systems already in development that are trying to combine and be flexible on the table so to have some very interactive options and to get the best information possible. As I say, these are very quickly evolving technologies, so we need to wait and see in which direction they are going. But in any case, IVUS is going to react giving much better quality, a larger bandwidth spectrum, higher frequency, so more detailed information.

Q: And Virtual Histology also?
Dr. Guagliumi: Virtual Histology, yes -- Virtual Histology is one name, perhaps tissue characterization is another name. The concept is the same. So with IVUS, the limitation was that we were analyzing just one single parameter: the intensity of the signal. Then we went to the analysis of the spectrum and that produced Virtual Histology.

For us in OCT it was the same. We started with intensity in the previous generation of OCT; now we are using what is called spectral domain OCT that is going to have the spectrum of the light that is millions of times superior to the spectrum of the sound.

Q: Do you foresee OCT technology impacting the use of histology or pathology in pre-clinical work?
Dr. Guagliumi: We are just transferring this technology to pre-clinical use. For example, in pre-clinical when you wanted to see a stent’s response in let’s say two weeks, you needed to take a group of treated animals and then sacrifice them to inspect the results using histology. But you will then need to see results at 30 days or six months, so you have to go back and restart the study with another series of animals. If we can gain the time dimension, to have a reliable technology (OCT) that is in line with the histology, that doesn’t require having to restart and sacrifice animals each time, you can really see what is going on over time. The need that we have today, one of the most important, is to have pathologic animals that are going to mimic what now we know is going to happen in humans, and use this technology to get the most appropriate device as soon as possible.

Q: What are the future trials that will be coming out?
Dr. Guagliumi: We have already finished OCTAMI -- OCTAMI is just to test if the uniformity of the response in zotarolimus is the same in acute myocardial infarction as it was for elective indications in ODESSA. Because, even if it's off-label, the setting is different and we wanted just to see if the response was going to be the same.

We have finished the enrollment and finished the follow-up. It's the identical platform but one is DES and the other bare metal stents.

    Ospedali Riuniti di Bergamo
Ospedali Riuniti di Bergamo

The second study that we are finishing shortly and perhaps will be submitted to and presented at TCT, is the new ablumenal polymer, JACTAX vs. TAXUS. LabCoat was recently acquired by Boston Scientific, and it is not just a polymer wrapping around the entire stent, but is only ablumenal, on the outside of the stent. Also, in terms of thickness, the polymer is just one micron thick, compared to 18 microns in TAXUS. So we are finishing the study: the primary endpoint is OCT and, of course, IVUS. So it is a comparison with the control arm, but in this case the control arm is the first generation of the stent. So this is the first prospective randomized study comparing a new stent generation with the first generation using OCT as a primary endpoint -- and the name is OCTDESI.

Q: At TCT 2007, OCT was hardly discussed. But at last year’s meeting there was a lot of talk about this. Would you say that the use of OCT is beginning to grow rapidly?
Dr. Guagliumi: Today, every day in Japan they are doing 450 OCT cases on average. There was nothing until one year ago. Of course I have a lot of friends in Japan and they are participating in our meetings as well.

The growth of OCT is explosive. I receive two or three manuscripts per week for peer review so as I said before, the preoccupation is for quality, because otherwise we're going to have conflicting data. And all the experts are starting to agree on some consensus documents talking about definition, technology -- it's a technology that’s growing so quickly, even in terms of the marketing impact, that we need to guarantee the quality of the data because it could be a serious problem. But many people are joined; we are having fluent conversation with our Japanese friends, U.S. friends and the Europeans; it's probably helping to quickly get an enormous quantity of data, yet at the same time guarantee a correct analysis.

This interview was conducted in April 2009 by Burt Cohen of Angioplasty.Org.