Intravascular Ultrasound in the Cath Lab: A powerful and underutilized tool to improve patient outcomes

Have the issues with drug-eluting stents and stent thrombosis been a wake-up call?
It has been a wake-up call. Of course, early on when this issue came up, all the focus was on the stent technology itself. People talked about how maybe we shouldn’t be putting drugs on stents. There was not much discussion about how the issue is perhaps not the stent technology, but in part due to how we are deploying these stents. From my perspective, drug-eluting stents (DES) clearly require more work to fully expand than the thinner bare-metal stents. That’s just common sense. DES have thicker strut material, and the struts are coated with a polymer, further restricting their ability to expand, and then of course, there is a deliberately delayed healing process with the medication. I became interested in DES expansion very early on, within a month or two of starting to deploy these stents. We saw some thrombosis issues and began to start using IVUS on DES patients to see if we could learn something that might explain these events. What we immediately saw is that even though the stents look good angiographically, IVUS showed many of the stents were underexpanded and in some cases, the struts weren’t properly apposed. I discovered you have to work hard with non-compliant balloons at high pressures to properly deploy DES. Recent work by Dr. Lowell Satler and others at the Washington Heart Center indicate this to be true. Interestingly, in our experience, once we started looking at DES with IVUS and working a little harder to get them expanded, we quit seeing the thrombosis issues. We haven’t published that data, but it certainly was anecdotally evident in our own practice. I hope that in the future we will see some data that will verify the value of using IVUS in DES procedures. The ADOPT DES stent registry, with Dr. Gregg Stone and Dr. Bruce Brodie as the national co-principal investigators, is just getting started and will be looking at this very issue.

Will any of the next-generation DES perhaps mitigate this problem, with their thinner-strut stents?

I am hoping that will be the case. I have had a limited experience with newer generation stents as an investigator in a Boston Scientific-sponsored trial called PERSEUS (A Prospective Evaluation in a Randomized Trial of the Safety and Efficacy of the Use of the Taxus^® Element^™ Paclitaxel-Eluting Coronary Stent System for the Treatment of De Novo Coronary Artery Lesions), with their third-generation stent, which is exceedingly thin and deliverable. Hopefully those interventionalists deploying DES without the benefit of IVUS and who are just using moderate pressures with delivery balloons will see that the next-generation DES are being better expanded. However, I don’t think we have seen enough data on the newer stents yet. Obviously the trials that get these stents approved by the FDA look good, but most of them are done with relatively simple lesions and in a relatively small number of patients. You don’t have the statistical power to determine very minor differences of rare events like stent thrombosis. As the experience progresses and people use the next-generation DES in more difficult anatomy, I think time will disclose those answers.

What are some of the patient or procedure characteristics that should prod a physician to use IVUS?
Let me disclose right at the beginning that in my personal practice, my IVUS use is at about 90%. If I am going to put in a DES, I want to be very certain in my mind that I have done my best to expand it. Do I believe every patient who receives a DES has to have IVUS? I think I could make an argument for it, but I probably don’t have the data at this point to make that determination.
There are at least three scenarios which clearly ought to include the use of IVUS. First, any patient with in-stent restenosis, because you don’t know the cause of in-stent restenosis by angiography, you only know the vessel is narrowed. In about a third of the vessels I see with in-stent restenosis, the stenosis isn’t from tissue growth or neo-intimal hyperplasia, it is a focally underexpanded stent. These patients don’t need a new DES; what they need is that stent to be properly expanded. I have seen many cases where there is absolutely no neo-intimal tissue at all. Obviously if there is extensive, diffuse neo-intimal growth that is the cause of that in-stent restenosis, a DES or a different DES than what they received before is probably the answer. So in-stent restenosis is one scenario that cries out for IVUS. Second, for any angiogram you are having difficulty understanding, whether you are uncertain if there is thrombus, or you are just having difficulty assessing the vessel size or the severity of the lesion, I would advocate IVUS. You are likely to get a definitive answer with IVUS and I would say parenthetically that virtually all angiographers, no matter how experienced, routinely underestimate the size of vessels when they rely on angiography alone. It’s surprising how often you are thinking of putting in a 2.75 or 3.0 stent, then you do actual measurements and it’s a 3.5 vessel. Most studies show that most angiographers underestimate the vessel size, by at least 0.5 mm or even up to .75 mm. For any unanswered question angiographically, I would advocate IVUS. Third, I’d advocate IVUS for complicated lesion subsets. For example, bifurcation lesions, ostial lesions, left main lesions — essentially those higher-risk, complex interventions where we know, even with DES technology, that you are likely to see higher rates of restenosis or a higher risk of stent thrombosis.
To my knowledge, national use of IVUS in this country is somewhere between 5%-15%, which in my mind is way too low. The Japanese are at about 70%. Depending on the types of cases people are doing, anywhere between a 30-50% level of use is a more appropriate number.

What have been some of the technological developments that have made IVUS easier to use?

To me, the most critical is integration into the lab. Any interventionalist may have a natural resistance to bringing in a standalone device, whether a Rotablator, an AngioJet, etc., where you have to drag another piece of equipment that takes up space into the room, then have personnel turn it on and run it. It is an understandable and very natural impediment to using any technology, and it obviously takes a lot of time. Integrating IVUS into the cath lab, where it is always on and ready, means IVUS use is just a matter of grabbing an extra catheter and turning on a button. Integration is the most important development we have seen in the 14 years since I have been doing IVUS. We have not had this absolute luxury of integrated equipment in cath labs until the last year or two. It has certainly made the procedure much more pleasant for me and for my staff. It’s quicker and has dramatically reduced the resistance staff have to doing IVUS. Times in the past, you could almost hear people’s eyes roll around in their heads when you’d say, ‘Well, bring in the IVUS machine, let’s take a look at that’ They just knew it was going to be an extra 10 or 15 minute process. So the level of staff resistance, which I think in many cases has affected interventionalists’ judgment on whether to use it in a procedure, really evaporated in our lab. Our staff now often perceives IVUS as a time-saver. We have timed it and it takes about a minute and a half to do a baseline IVUS run in our lab. With that image, I can make a definitive judgment on the size of the stent I need and whether I will need to perhaps post-dilate with a larger balloon to upsize the stent. In some cases, I will look at a lesion and conclude that it doesn’t even need to be done at all. With integration, IVUS doesn’t prolong the procedure and may ultimately be a time-saver. I can even predict with a high degree of certainty from the baseline IVUS whether a lesion will require a pre-dilatation to get a good result.

Virtual histology (VH) is a fascinating advancement; I think it is an advancement. At this point, in routine practice, it doesn’t have a clear-cut niche or indication. I use it, it’s part of the system we have, so I routinely look at IVUS with VH. I do gain a lot of insight into the nature of the lesion. Sometimes I make decisions to extend extra stents in adjacent disease. Sometimes VH points to a culprit it area of disease proximal to the area of the most severe stenosis. Recent studies indicate lesions with high necrotic core content appear to be more likely to embolize debris, and it might lead me to a decision to place a distal protection device. At this point we are still learning about VH. The PROSPECT trial should provide insight into the ability of VH to detect vulnerable plaque; this trial is entering the third year of follow up. If VH proves to be useful in vulnerable plaque detection, we may one day be able treat lesions before they cause potentially catastrophic clinical sequlae. At this point, however, we are still learning about VH.
As you noted, is important to place the ends of the stent in a normal, undiseased part of the artery if possible.

Is it possible to do this with angiography in most cases?
Most physicians want to try to extend their stents to normal-to-normal tissue if possible, particularly with DES. With bare-metal stents, we paid a big price — the longer the stent we used, the higher the rate of restenosis. Unlike bare-metal stents, DES do not have the big change in restenosis rates according to stent length, but we do have the issue of geographical miss with DES. If you stent a fairly short segment of vessel and are leaving the ends of the stent in diseased segments, there most likely will be a price to pay with edge restenosis. My own approach with DES has been to stent from normal-to-normal tissue when possible. Most interventionalists try to do just that, but if they are using angiography to determine the normal tissue, we know that angiography is not very sensitive for detection of mild-to-moderate disease. IVUS is much better, so if you are using angiography to determine where your normal reference vessel is, in many instances, you are simply mistaken. The reason why angiography is so insensitive is that as plaque deposits in the vessels, the vessel remodels or expands to accommodate that plaque. Remodeling is a well-observed phenomenon that we have known about for over 50 years, and we can see it with IVUS. If I am using IVUS to determine normal tissue, the determination will be accurate. As a result, the rate of stent edge restenosis and also in-segment restenosis, defined as within 5 mm of the stent, will be reduced. Again, however, that’s observational data from my own experience. Hopefully we will have stent registries that start looking at some of these issues in the future.

What are the limitations of IVUS?
The main limitation is that IVUS is a catheter and there are certain vessels the catheter can’t go down: tortuous vessels and highly angulated take-offs of circumflex vessels are often difficult to ultrasound, for example. The IVUS catheter often will not pass through highly calcified vessels. So there are technical limitations on vessels you can IVUS. Of course, it is an invasive procedure with the attendant risks. People who are not interventionalists, i.e., non-interventional physicians, should not be using IVUS because there are rare vessel dissections that can occur with IVUS. Although, interestingly, I don’t think I’ve ever seen a severe vessel dissection from IVUS use in the 14 years that I have been using this technology. The main morbidity is intracoronary spasm, which is easily preventable with ample intracoronary nitroglycerin. So the two main limitations are vessel anatomy — some anatomy is not suitable for IVUS — and that it is an invasive procedure, which limits the application of its use to some extent.
An additional limitation is that many interventionalists are not trained in intracoronary ultrasound. They simply do not have a good understanding of how to interpret the images, how to measure properly, how to interpret the measurements properly and so forth. I have now been giving training courses for several years to interventionalists not using a lot of IVUS. The last thing most experienced interventionalists want to admit to anyone is that they don’t really know how to use IVUS correctly. But I find when I get people into my courses, by the questions they ask, by their own admission or by just asking questions to them myself, that many, many interventionalists don’t clearly understand how to interpret and use IVUS. Fortunately, it is obviously a very correctable limitation. One of my concerns is that many academic training centers actually have relatively low IVUS usages. It is not everywhere, but I do know of training programs where the use of IVUS is very low. How can you train fellows to be expert in IVUS if you’re not using it in the first place?

Do you find that experienced staff are helpful in IVUS use?
We do have some staff who are experienced and who have been to IVUS courses. Some of our staff are extremely good at both the mechanics (the “buttonology”), and measuring and interpreting IVUS images. If an interventionalist who is not a frequent IVUS user is working with a staff member who is very comfortable using IVUS, we sometimes see that the physician is more likely to IVUS. Sometimes a technologist will suggest to the physician, ‘Why don’ you IVUS that?’But obviously if that tech is very uncomfortable with IVUS, they will not make that recommendation. Labs with IVUS should definitely make an attempt to train as many people as they can to be adept at both doing IVUS and understanding the image they see.