Latest / Behind The Knife: The Surgery Podcast / Journal Review in Vascular Surgery: Disrupt PAD III: Intravascular Lithotripsy for Calcified Peripheral Artery Disease
Transcript
- BTK_disrupt_padIII === [00:00:00] Welcome back to Behind the Knife. Today, I have the group from the University of Michigan. This includes Robert Beaulieu, Frank Davis, Carolyn Judge, and myself, Andrew Huang. Today, we're doing a deep dive into two publications from the Disrupt PAD III program, a program that has fundamentally shaped how we think about treating calcified peripheral artery disease using intravascular lithotripsy. We'll be covering the randomized controlled trial prov-- published by Tepe and colleagues in JACC, um, back in twenty twenty-one, and the final analysis of the observational studies pub-published by Armstrong colleagues in the Journal of Endovascular Therapy in twenty twenty-six. So let's start here first with the clinical problem. Why does calcium matter so much in peripheral arterial disease? Well, Andrew, that's, that's a really great question, something that we at the vascular surgery group and vascular surgeons across the country really struggle with for treating. Um, it's really a foundational understanding of why these studies were needed. Indeed, vascular calcification in the femoropopliteal [00:01:00] segment, but actually across the entirety of the peripheral vascular disease, is incredibly common, especially in our aging population, as well as in patients with diabetes and chronic kidney disease. The problem is that calcium makes everything harder. Standard balloon angioplasty in a heavily calcified vessel is often results in inadequate luminal expansion of the blood vessel, higher rates of blood vessel dissection, resulting in the need for bailout stenting. And we know that stenting the femoropopliteal segment is not a benign intervention. In-stent restenosis rates at one year have been reported anywhere from nineteen to forty percent, and those rates are even worse in calcified arteries. So you're telling me that the clinical question for vessel preparation really is: Can we do better than a standard balloon angioplasty, uh, before definitive treatment? Exactly. Exactly. And I think that's the big thing that we at vascular surgery right now before these studies have really struggled with how best to treat the calcified vessel. And the concept here is vepal prep-- vessel preparation. The idea that you can modify the [00:02:00] calcium, improve the luminal compliance of the blood vessel before you apply a more definitive therapy, whether that's a drug-coated balloon or if needed a stent And historically, vessel preparation have included angioplasty, cutting balloons, or even atherectomy, but each has both positives and negatives to choosing those treatment modalities. More recently, intravascular lithotripsy, or IVL, emerged as a novel approach to this problem. The shockwave system uses electrohydraulic emitters mounted inside a standard angioplasty balloon that deliver pulsatile sonic pressure waves. These waves create microfractures, both superficial and deep, within the calcium deposits that are located in the blood vessel. In essence, they injure that calcified deposit and allow it to be more compliant to different, different therapies. The objective is to crack the calcium at lower balloon inflation pressures, which should theoretically favor fewer dissections, lead to less trauma to the blood vessel, and ultimately lead to a better long-term outcome for the patients. And I [00:03:00] think, Frank, it's so important that you mention this can happen in almost any vascular bed. We're gonna be talking about the DISRUPT-PAD trials today. Mm-hmm. But there are a series of trials that involve coronary artery disease, or the DISRUPT-CAD trials. It's really a testament to how vascular calcifications, and specifically medial arterial calcification that we see so commonly in diabetes and end-stage renal disease, really affects vessels and impacts the way that we can ultimately treat them. Because I think the idea of you just expand a balloon and no matter what's causing the narrowing, be it soft atheroma, or if it's neointimal hyperplasia from scarring, or if it's calcification, that they're just gonna respond the same, is pretty much out the window. And now we've got a tailored treatment paradigm based on the type of narrowing, or at least what's causing the narrowing. Good points. [00:04:00] So going off that, for peripheral artery disease, before the DISRUPT-PAD th- th- three trial, what evidence did we really have about intravascular lithotripsy? Yeah, I think, I think overall it was limited. First we had a European pre-market single arm DISRUPT-PAD one feasibility study, which was the precursor to this given study, and that only consisted of 35 patients, which had symptomatic femoral popliteal disease, and which achieved in their pre-study at only 35 patients 100% procedural success and only 23% residual stenosis Then there was the Disrupt PAD-2, which was the further progression of that cohort, and was a prospective single-arm multi-centered study. And this study was 12 months with follow-up of 60 patients, again, treating that femoral popliteal disease segment. This study demonstrated that similar results, there was 24% residual stenosis, and the one year clinically driven target [00:05:00] vessel revascularization rate was 8.6% with optimal technique. However, they were all single-arm studies without randomized comparison. So the big gap was how would this perform in a fully randomized trial, and that's what Disrupt PAD-3 or the study we're talking about today sought to answer. Yeah. So let's talk about the Disrupt PAD-3 program then. This was two parallel studies running simultaneously, the randomized controlled trial arm and the observational trial arm. So the randomized controlled trial, the TC-- uh, RCT arm, the-- had a very specific objective, and this was to test the hypothesis that vessel preparation with intravascular lithotripsy prior to definitive treatment with a drug-coated ballo- which that definitive treatment would be a drug-coated balloon or provisional stenting, would result in superior procedural success compared to just standard angioplasty followed by that same definitive treatment. The endpoint for this study was procedural success, which this study defined as no, uh, no more [00:06:00] than a residual stenosis of 30% or with-- a-a- and not having a flow limiting dissection, which was all assessed by independent core laboratory, which was looking at these angiographic images. Importantly, this was perform-- All this assessment was done after the vessel preparation, but before the definitive treatment that's either with the DCB or the stent to make sure that any of the endpoints that were met were after the, the vessel preparation itself, itself, which is a novel endpoint design. The powered secondary endpoint was a 12, uh, month primary patency. The observational arm of the study, um, that was published in twenty twenty-six by Armstrong and, and, um, had a different but complementary objective. It was designed to evaluate the real-world acute performance of intravascular lithotripsy over a much broader patient population and across multiple vascular beds, not just the femoral popliteal segment as in the, uh, randomized controlled trial arm. This [00:07:00] observational arm included the iliac, common femoral, and infrapopliteal, uh, arteries as well. However, because it's an observational study, it's just a single arm registry, so there was no comparator, but the true value that it provides is the scale and complexity of the patients enrolled. Yeah, I mean, I, I think too if you look at it, this is how a lot of trial patterns for new devices go these days. I mean, you have a feasibility trial, you have a safety trial, and then you really do have a clinically driven endpoint trial, which is what DISRUPT-PAD-III was. And, you know, kudos to practicing physicians for saying, "Those data aren't enough, and we need to know in a real world scenario how do patients respond to this therapy you're suggesting." And so it's actually quite common. I don't think this is an abnormality that trials looking at getting devices approved now have large registry data available afterwards to say, "Once we release this to the wild, so to speak, how does it truly behave?" And so I'm [00:08:00] super excited for us to talk about this Armstrong paper more, as it gives us a real chance to dive into some of those real-world outcomes that may actually drive our clinical practice more. Let's walk through the study designs then in detail. Let's start with the RCT and then the observational study. Sure. Uh, the randomized trial was a global single-blind multi-center trial conducted at forty-five centers across Austria, Germany, New Zealand, and the United States. Uh, patients were eligible if they had moderate or severe calcification in the femoral popliteal segment and were symptomatic chronic PAD Rutherford classes II through IV. They randomized three hundred and six patients, one to one, a hundred and fifty-three in the lithotripsy group, and a hundred and fifty-three in the plain balloon angioplasty group. Uh, both groups then received definitive treatment, uh, with the drug-coated balloon with provisional stenting allowed if needed. Patients were blinded to their treatment allocation, and the duplex ultrasound core laboratory readers were also blinded. It's important to note that the investigators and the cath lab [00:09:00] staff were not blinded. All analyses followed intention to treat principles. Core laboratory adjudication was used for the primary endpoint, and a clinical events communi-- uh, committee adjudicated adverse events And what about the observational study? The observational study itself was a prospective multicenter single arm registry without randomization or comparator. Uh, so it enrolled 1,300 patients, uh, with 1,600 and change lesions across 30 global sites between November of 2017 and June of 2021. Their inclusion criteria were a little bit broader. They enrolled patients with claudication or critical limb threatening ischemia with at least moderate calcification, but there was no restriction on the vascular bed. So you had femoropopliteal lesions making up about 61%, iliac lesions at about 16, and common femorals, uh, at right about 11%, and then infrapopliteal, uh, lesions at 13%. So the key here is the complexity of the [00:10:00] populations. Um, nearly a third, uh, 31% or so were chronic total occlusion, and about 19% of those were long lesions coming in at fifteen centimeters or greater. The mean lesion length was nearly 100 millimeters, right about 94 millimeters to be exact, and 91% of those lesions were core lab, uh, evaluable, which is a high rate for a registry of this size. Um, they also performed multivariable analysis to identify independent predictors of procedural outcomes, which added an important, um, analytical barrier, uh, layer rather, uh, between beyond simple descriptive, uh, statistics. Yeah. I think it's super important that you mention in the study design what the Rutherford classifications are for patients because, you know, and this is different than the Rutherford that you may think about when you're getting ready to take your general surgery boards, which is more geared towards acute limb ischemia. This is the chronic limb ischemia Rutherford classification, which actually has six. And so these two, three, [00:11:00] and-- or sorry, two and three are sort of in the claudication category still. It's m- moderate and severe claudication. And once you start getting up to four, you get a little bit more of the rest pain aspect of it, which we would probably put definitively in the CLTI group or the chronic limb threatening ischemia. That short distance claudication is always kind of the gray zone, and, and many of the trials it would be kind of the Rutherford three. In the, the Disrupt PAD three trial that was originally released, the vast majority, like 95% of the patients were the Rutherford two and three classification patients. And so as we start to talk about the results here and coming up, I think it's gonna be important for us to remember are there differences in who got treated in the approval trial, the Disrupt PAD three original trial, and those that were seen in the observational study? Because did the world take the information it gained from the, the [00:12:00] Disrupt PAD three trial originally and then expand its use beyond maybe the study population that was originally studied? So let's talk about the results. We can start with the randomized controlled trial and then with maybe some of the headline findings. Yeah. So the primary endpoint, which was defined as procedural success, which was, um, measured by a blinded core lab, w- met statistical significance. So intravascular lithotripsy achieved procedural success in 65.8% of the lesions compared to only 50% with plain old balloon angioplasty, and that was a p-value of .01. Now, 66% may not seem like it's that impressive, but it's really important to remember that this means the residual stenosis as judged by an independent body was 30% or less, and that group was blinded to what the treatment algorithm was before. It's also really important to remember that as the trial was designed, as Carolyn just told [00:13:00] us, it was actually designed with this idea of let's see what the lesion looks like after vessel prep but before definitive treatment. So it was looked at in patients who had only gotten the lithotripsy and not yet gotten the drug-coated balloon afterwards, which is the definitive treatment that was most typically used in this group When you look at the site-reported success, intravascular lithotripsy was about ninety-f- ninety percent, excuse me, versus sixty-five percent for PTA. The discrepancy between site and core lab assessment is kind of one of these well-known things in peripheral artery trials and actually, you know, strengthens the argument that we should have out there that having a third-party core lab adjudicated result makes a lot of sense. What about the secondary procedural outcomes then? Yeah. I actually really like this part of the story because, um, the intravascular lithotripsy, for those that use it, know that it is a much smaller [00:14:00] pressure to which you inflate the balloon. And so the most typical pressure is around six atmospheres, and this compares with over eleven atmospheres for plain balloon angioplasty. And it may not seem like that's a big deal, but really I think that's what impacted the next part, and that's the much lower rate of dissection following balloon angioplasty in the lithotripsy group. So that was eighteen point five percent versus thirty-two point three percent, and that had a statisti- statistically significant P value. When you look at dissections that are flow-limiting... So I think a lot of us will do these angioplasties, and you may see a small dissection flap, and if it's non-mobile and non flow-limiting, you might just sit on it. But the flow-limiting dissections are like, "Hey, we're probably coming back either tonight or soon because this is gonna take down our repair." That rate was only one point four percent in IVL and six point eight percent in plain balloon angioplasty, and again, that had a statistically significant [00:15:00] P value. And so what do you do most of the time when you see a dissection or residual stenosis that you think's gonna ultimately result in a compromised result? Well, then you often will do this provisional stenting. And as we've talked about before, the importance of measuring the rates of provisional stenting is getting a sense of, okay, this could either fail, 'cause now there's a stent in place, or does it take away a future option 'cause you have to stent beyond the length of just the lesion? And that rate was four point six percent in the IVL group compared to eighteen point three percent in the patients who had plain balloon angioplasty. So you've talked about dissection and a lot of these secondary endpoints. What about the rest of the safety data? Yeah. I mean, I, I think, you know, the good news is this looks to be a pretty safe treatment profile. And the major adverse events at 30 days were essentially 0% in the IVL group. And the good news is even in the POBA group, it was only 1.3%. So luckily, I think as a group, we've gotten pretty good at selecting the right [00:16:00] patients to do these procedures on, and we've m- therefore been able to really lower some of the major adverse limb events. The clinically driven target revascularization rate was i- um, identical between the groups, and that was 0.7%. That's a really important outcome because clinically driven is different from radiographically driven target vascular revascularization. So these are patients who end up having a problem and needing it fixed within the treat- the, the study period, and that's only 0.7%. So that's, that's pretty awesome. The one trade-off was like procedural time, and that's longer in the IVL group. Um, I think that's because you're doing more balloon inflations, and you leave it inflated for this, this pulses. The generator itself, so you hook it up to this electric generator that's gotta generate the pulses. And so the pulse frequency at which a generator can generate those pulses has a huge impact. And since this trial, the company's more contemporary generator generates about twice as many pulses. So it's 300 [00:17:00] pulse rather than a, a 180 pulse generator, so. Then what about longer term patency? Yeah. Um, the 12-month primary patency data, uh, which was the powered secondary endpoint, was reported subsequently and showed that lithotripsy achieved significantly better primary patency of 80% versus 68% for the POBA group, again, with a statistically significant p-value. And at two years, that separation continued to widen at 70% and 51% respectively, again, with a statistically significant p-value at 0.003. Um, it's worth noting that, uh, acute angioplasty failure requiring stent placement during the index procedure was counted as primary patency loss, which does impact the interpretation of those numbers. Um, now let's turn to the observational study. What did Armstrong colleagues find in that much larger cohort and with much more diverse vascular beds? Right. The, the headline finding is remarkable consistency. Um, across over thirteen hundred patients and again sixteen hundred [00:18:00] lesions in four different vascular beds, uh, the final residual stenosis was twenty-three point eight percent, which is almost identical to the twenty-four percent residual stenosis seen in the randomized trial and in every prior DISRUPT-PAD trial. Um, the serious angiographic complication rate was just shy of one percent at point nine. Uh, and there were zero abrupt con- closures, zero distal embolization events, zero no flow or slow flow events, and zero thrombotic, uh, events. So in a registry of nearly fourteen hundred patients with heavily calcified arteries, that is a remarkable, uh, safety signal. Yeah. I, I think this is like what we were talking about earlier. One of the really important things to look at is the difference in this study population. So like we said earlier, in DISRUPT-PAD III, it was like ninety-five percent claudicants. In the Armstrong paper, it was like eighty percent CLTI group. So we're talking Rutherford stage four and five. And so despite [00:19:00] different vascular bed with potentially higher risk, as well as patients with worse limb outcomes, so if they had rest pain or they actually have wounds on their limb, sort of limb presentations we'll say, the outcomes were the same. They actually did really well. So this is really encouraging finding to take-- to think that, okay, I wouldn't call it off-label use, but people used their lithotripsy for more than what was studied in the original DISRUPT-PAD III trial, and the results continued to be encouraging. What about the multivariable analysis? What other factors and variables contributed to the outcome that they determined? Yeah. So I, I think this was one of the most valuable contributions of this paper. Uh, they identified several independent predictors of achieving that residual stenosis of thirty percent or less. Favorable predictors included female sex with an odds ratio of one point eight five, age of seventy-five or younger with an odds ratio of one point six, and critically, an IVL balloon to [00:20:00] artery ratio of one or greater with an odds ratio of one point five. Um, that last finding is very actionable, um, as it tells operators that proper balloon sizing matters. If you're undersizing your lithotripsy balloon, you're leaving efficacy on the table. Um, on the flip side, lesions, uh, length of fifteen centimeters or greater was a negative predictor of procedural success, uh, with an odds ratio of point three eight, and chronic total occlusions also had lower odds of achieving that residual stenosis target, um, with an odds ratio of point six four. Lesion length of fifteen centimeters or greater was also the strongest predictor of procedural complications overall with an odds ratio over sixteen. Um, so the takeaway is really long, complex lesions remain challenging, um, even with lithotripsy. Okay. So let's put-- We have two studies here, one observational, one randomized control. Let's put it all in context. What do these two papers tell us about the purpose and the use that, [00:21:00] of IVL or intravesical lithotripsy in PAD? Yeah, no, great question, Andrew, and I think that's kind of like the take-home thing of these two studies, but also the prior Disrupt PAD studies too. I think overall the combined message is very powerful. The randomized controlled trial gives us the rigor-- a rigorous investigation, excuse me, proving that in a well-designed core lab adjudicated randomized trial, that IVL is superior to plain old balloon angioplasty in a vessel preparation strategy. It achieves better lumenal gain, fewer dissections, and dramatically less bailout stenting. Uh, added on top of that is the observational study, which gives us the generalizability of this technology outside just this infrafemoral popliteal segment, showing that these results also hold true in a broad spectrum real-world population, um, in complex patients with CTOs, long lesions, as well as a scale of other aspects prior to IVL studies. The, and I think it was also really, really interesting that the consistency of IVL [00:22:00] across different studies was approximately 24% residual stenosis across all disrupt PAD studies. From the fewer to 35%, or excuse me, fewer to 35 patient feasibility study all the way up to the 1,373% patient registry. It's quite superior that a device can have that much consistency in its results. It suggests that IVL is a very predictable and reproducible treatment in its given effect. So Frank, I don't want to put you on the spot- Yeah ... but you see these data. Yeah. Are you doing IVL for everybody? I- How do you interpret them? Yeah, I, I'm an IVL adopter. For those patients who have calcified lesions in their iliac common fe- or excuse me, iliac femoropopliteal segment, I think IVL is definitely much, much better than percutaneous balloon angioplasty. I'm not a fan of trying to... I just want to limit my number of stenting or bailout stenting in there because I think the long-term ramification on patients is profound. So if I can adopt a technology like IVL that has a good reproducible result, decreased risk of bailout stenting, and has good long-term patency, at least at one [00:23:00] year, the 12 months that they show here, I'm all for it. And you, you backing up with the DCB or drug-coated balloon afterwards or- Yeah. Exactly, because that's how, that's how it was designed in the study. And I definitely think there's added benefit to having the drug-coated balloon on top of the IVL. The IVL does the adequate vessel preparation, but it also, you have to back it up with a DCB from that standpoint. All right. And then maybe a little off the cuff one, not related to the trials, but related to IVL. Uh-oh. So are you... How, how are you sizing it? Are you basing it on preoperative CT? Are you looking at angiographic sizing? Yeah. Are you putting in IVUS for these patients? What do you do? Yeah, so I usually, um, do pre- or intraoperative angiogram to size it, and then if I'm at all concerned, if there's wide discrepancy between what I actually size it on my CT scan, uh, preoperatively, then I run an IVUS down there too. Yeah. Yeah. There's nothing quite that'll generate so many eye rolls from your techs than to ask them to pull out IVUS- Yeah ... and IVL, and then go get another balloon for you out of another room. It- But I totally agree with you. If you have any question about it, run an [00:24:00] IVUS down there. I mean, it's probably one of the take-home messages from most of the vascular conferences recently is we're probably under sizing things and- Yeah leaving, as Carolyn put it, efficacy on the table. Um, but it's a tall order to IVUS everybody. It is, yeah. I mean, I think if you look at a lot of the cardiology based literature, I- IVUS is way more profoundly utilized in their field. And I think there's both positive and negatives to it, but I think vascular surgery across the board could benefit from further utilization of IVUS in their sizing. I think also a discussion of cost too in that opening an I- an IVUS catheter isn't free when you're opening that and- You know, making some device companies happy. Well, speaking about cost, what about atherectomy? Oh That's another one of these devices that we sometimes use. You know, uh, you know, it's a calcium modifier. Well, how do you put that into context? I, I think I, I have my own personal opinions about atherectomy, but I'll say it to the broader context of the, the data. So I think the big elephant in the room, and the honest answer is we don't have the head-to-head comparison between IVL and atherectomy right now. The Disrupt PAD III trial only [00:25:00] compared IVL to plain old balloon angioplasty, not atherectomy. So we really need additional randomized trials to say, is IVL better or worse than atherectomy from that standpoint. So what about the, the whole thought of leave nothing behind- Mm-hmm ... and just, and leaving future options for yourself? Yeah, so I think the central theme of the Disrupt PAD per- III program is the idea is that by optimizing vessel, vessel preparation with IVL, you can treat then with a drug-coated balloon alone and leave nothing behind in terms of stenting or a met- metabolic, or excuse me, a metallic implant. In the randomized control trial, only 4.6% of patients with IVL needed a stent compared to 18.3% of patients with plain old balloon angioplasty. And indeed, I think that's a really important because the femoral popliteal stents are subject to kind of unique biomechanical stress, especially as you approach that popliteal artery with re- flexion, compression, torsion from that standpoint. Additionally, in-stent restenosis is a major [00:26:00] clinical problem, and reducing the need to stenting with using IVL can prevent the need for stent and prevent those complications themselves. So let's wrap up this whole conversation here. If you, we can direct it to our staff in the room, what would you say would be the takeaway points from these two papers? Yeah. I mean, I think both from the Disrupt PAD III trial and then Armstrong's follow-up registry data can pretty clearly say that in the right setting, that IVL is superior to plain balloon angioplasty as a vessel preparation strategy, especially in calcified femoral popliteal disease It's level one evidence from a well-designed RCT that had, as we talked about, core lab adjudicator results. If you're treating a calcified femoropopliteal lesion and planning on using drug-coated balloon, I think you should lean towards preparation with IVL as opposed to plain balloon angioplasty. So, I mean, that'd be point one. I think that's supported by point two, which is [00:27:00] using IVL actually also reduces the need for bailout stenting, typically because of the lower flow-limiting dissection rates and dis-- and residual stenosis rates. We really try to not put a stent in place when possible because you're talking about, first of all, typically intervening on areas with high flexion and high movement arteries, both at the adductor hiatus and sometimes even behind the knee. And so it's really nice not to have a stent there. And it also takes away future options when you're stenting because you do, as we've talked about, have to extend your stent beyond just the area of the narrowing into the normal vessel on each side. And I think this is sort of bolstered by the idea that we've now got multiple studies. You got PAD one through PAD three, the disrupt trials, and then the observational studies that show the safety and effectiveness of this Back to the point, and I'm gonna kick this one over to Carolyn too, you know, when we talk about sizing, we're looking at [00:28:00] both the size of the vessel and the type of, you know, lesion that's there. When you are preparing for your cases that you're gonna run your plans by the attendings, do you tend to look at what the lesion itself looks like, or are you just looking at luminal difference between an open artery and a narrowed artery? Yeah, absolutely looking at the lesion. Um, I think my takeaway from this data is really that in these heavy circumferentially calcified, uh, arteries, that's really a factor, um, in considering lithotripsy. Yeah, absolutely. I mean, and I, I, I think we're really expanding through the observational trial results, which would kind of be the final takeaway. We're probably expanding what was seen in PAD3 to other vascular beds and showing that this can be beneficial. Um, you know, you're not gonna have any trial support for carotids and renals and SMAs just yet, but [00:29:00] I'd be lying if I said I hadn't talked to people that had used it in all of those locations. And so what I, what I think we're seeing is it's safe, doesn't seem to embolize things in the same way that we get worried about maybe other some calcified modifying technologies do, and seems to... I don't know if it's like, Frank, you're the science guy. Does the Paclitaxel get into the wall better because there's these disruptions or... I don't know what's going on there, but it seems to allow for these drug-coated balloons to do better than they even do on their own. Well, I think the, the studies and the authors of these studies, and even device companies will show that the, um, the intravascular lithotripsy creates microfractures, and the histological sections actually can prove that. So I, I truly do believe that the drug from the drug-coated balloon after you use intravascular lithotripsy has more efficacy in these calcified lesions compared to plain old balloon angio- angioplasty because of the depth of penetration of the drug. So I think, I mean, yes, they're, they, that's what they tout, but from a histological and a [00:30:00] scientific standpoint, they've shown it, too. So I think it's a good thing. Nice. All right, Andrew, I think before we start talking out of one end a little bit too much- ... maybe you want to give us the sign off. Thank you, everybody, for having us. Um, and that was a comprehensive review of the randomized control trial as well as the o- o- observational component of the Disrupt PAD3 trial. And dominate the day.